World Association for Sport Management Series

MARKETING ANALYSIS IN
SPORT BUSINESS
GLOBAL PERSPECTIVES

Edited by Kevin K. Byon, Brian H. Yim,

and James J. Zhang
Marketing Analysis in Sport Business

This book explores sport marketing analytics, an essential and crucially impor-
tant aspect of contemporary sport business. Successful sport marketing begins
and ends with the consumer, so understanding the consumer experience is crit-
ical. Marketing Analysis in Sport Business demonstrates how rigorous analytical
procedures are the key to developing effective, evidence-based marketing prac-
tices that reflect real consumer needs.
Presenting cutting-edge case studies of sport marketing analytics in action,
the book explores topics such as digital communications, social media, digital
ticketing, event marketing, the economic impact of COVID-19, developing sport
brands, and conducting research with athletes and event participants. Written
by a team of authors from 15 countries, including Australia, China, France, Iran,
Italy, Japan, Mexico, New Zealand, Serbia, South Korea, Spain, Taiwan, the
United Kingdom, the United States, and Venezuela, the book offers insight from
a variety of cultural contexts and new perspectives on the global sport industry.
Marketing Analysis in Sport Business is illuminating reading for any advanced
student, researcher or professional working in sport business and management,
sport development, marketing, strategic management, or international business.

Kevin K. Byon is a Professor and the Doctoral Program Coordinator of the Sport
Management Program in the Department of Kinesiology at Indiana University
Bloomington (IUB), Bloomington, IN, USA. His primary research interest
involves exploring psychological and environmental variables affecting sport
consumer behaviors in an effort to advance theories and practices of sport mar-
keting and sport tourism. To date, Dr. Byon has authored or co-authored 120
research manuscripts and has served as an editorial board member of seven schol-
arly journals including the International Journal of Sports Marketing, Sponsorship
and Sport Marketing Quarterly. Dr. Byon was also the sport management section
editor of the Measurement in Physical Education and Exercise Science journal from
2010 to 2016.

Brian H. Yim is an Associate Professor of Sport Administration in the School

of Foundations, Leadership and Administration at Kent State University, Kent,
OH, USA. His overarching research interest lies in customer relationship mar-
keting with a particular focus on sport fan emotions in various contexts such
as professional sport, global sport, fantasy sport, and esports. To date, he has
authored or co-authored about 30 articles published in journals including Euro-
pean Sport Management Quarterly and the International Journal of Sports Market-
ing and Sponsorship. Dr. Yim is the Editor-in-Chief of the Asia Pacific Journal of
Applied Sport Sciences and serves as an editorial board member of the International
Journal of Sports Marketing and Sponsorship.

James J. Zhang is a Professor of Sport Management at the University of Geor-

gia, Athens, GA, USA. His primary research interests are applied measurement
and/or applied studies examining sport consumer and organizational behaviors.
He has previously been Editor of the International Journal of Sport Marketing and
Sponsorship, Sport Management Section Editor of the Measurement in Physical
Education and Exercise Science journal, and President of North American Society for
Sport Management (NASSM).
World Association for Sport Management Series
Series Editors:
Brenda G. Pitts, Georgia State University, USA
James J. Zhang, University of Georgia, USA

The World Association for Sport Management (WASM) was founded to facilitate
sport management research, teaching and learning excellence, and professional
practice, across every continent. The WASM book series is designed to support
those aims by presenting current research and scholarship, from well-established
and emerging scholars and practitioners, on sport management theory, policy and
practice. Books in the series will explore contemporary issues and key challenges
in sport management, and identify important new directions for research and
professional practice. Above all, the series aims to encourage and highlight the
development of international perspectives, international partnerships, and inter-
national best practice in sport management, recognizing the globalised nature of
the contemporary sport industry.

Available in this series:

International Sport Business Management

Issues and New Ideas
Edited by James J. Zhang, Brenda G. Pitts and Lauren M. Johnson

Sport Governance and Operations

Global Perspectives
Edited by James J. Zhang and Euisoo Kim

Sport Marketing in a Global Environment

Strategic Perspectives
Edited by Ruth M. Crabtree and James J. Zhang

Marketing Analysis in Sport Business

Global Perspectives
Edited by Kevin K. Byon, Brian H.Yim, and James J. Zhang

For more information about this series, please visit: https://www.routledge.com/

World-Association-for-Sport-Management-Series/book-series/WASM
Marketing Analysis in Sport
Business

Global Perspectives

Edited by
Kevin K. Byon, Brian H.Yim, and
James J. Zhang
First published 2023
by Routledge
4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
605 Third Avenue, New York, NY 10158
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2023 selection and editorial matter, World Association of Sport
Management (WASM); individual chapters, the contributors
The right of World Association of Sport Management (WASM) to be
identified as the author of this work has been asserted by them in
accordance with sections 77 and 78 of the Copyright, Designs and Patents
Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission in
writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record has been requested for this book
ISBN: 978-1-032-29875-7 (hbk)
ISBN: 978-1-032-29876-4 (pbk)
ISBN: 978-1-003-30247-6 (ebk)
DOI: 10.4324/9781003302476
Typeset in Goudy
by KnowledgeWorks Global Ltd.
Contents

List of Figures ix
List of Tables x
About This Book xiii
List of Contributorsxiv

1 Sport Marketing Analyses: An Introduction 1

KEVIN K. BYON, BRIAN H. YIM, HYUNSEOK SONG, AND JAMES J. ZHANG

2 Application of Structural Equation Model in

Sport Marketing Analyses 24
KEVIN K. BYON, BRIAN H. YIM, JUNYING AN, AND JAMES J. ZHANG

3 Disruptive Technologies on Sport Event Marketing:

New Centrality of Digital Communication 51
PATRIZIA ZAGNOLI AND ELENA RADICCHI

4 Challenges in Sport Marketing: A Systematic Review

of Digital Sport Marketing 70
SALVADOR ANGOSTO, ALEJANDRO LARA-BOCANEGRA,
M. ROCÍO BOHÓRQUEZ, JERÓNIMO GARCÍA-FERNÁNDEZ, AND
JAMES J. ZHANG

5 Digital Ticketing Utilization in High School Athletics:

The Role of Spectator Demographics 93
ARMIN A. MARQUEZ AND BETH A. CIANFRONE

6 Social Media and Stakeholder Relationships: Analyzing

an Online Communication Network in the Sport Industry 109
JUHA YOON, WONSEOK (ERIC) JANG, AND PAUL M. PEDERSEN
viii Contents

7 Influence of Micro-Celebrities on the Formulation of

Social Media Marketing Strategies 125
BENJAMIN MOLE, PAUL COOK, AND RUTH M CRABTREE

8 Examining Facebook Marketing Practice:

The Case of New Zealand Provincial Rugby 144
JASON COLE, ANDREW J. MARTIN, AND ASHLEIGH-JANE THOMPSON

9 Analyzing Sport Fan Facebook Pages: Influences of Post

Characteristics and Brand Attributes on Fan Engagement 163
SHANG-CHUN MA, CHING-HUNG CHANG, SHANG-MING MA,
WAN-CHEN CHEN, AND JAMES J. ZHANG

10 Marketing Obstacles Facing Obstacle Racing 182

JASON REESE

11 Congruity Perceptions in Endurance Event

Participant Consumers 199
JASON A. RICE, MARION E. HAMBRICK, AND T. CHRISTOPHER GREENWELL

12 Service Provider Selection with the Analytic

Hierarchy Process: An Approach to Sport Facility
Enterprization in China 218
YUNXIA DING AND YELENA Y. WANG

13 COVID-19 and Economic Impacts of Major Sport Events 242

LIDIJA PETROVIĆ, MIAAD GHASEMZADEH, AND CHRISTOPHER HAUTBOIS

14 Brand Attributes Scale for Professional Sport Teams:

Measuring Strength, Favorability, and Uniqueness of
Team Attributes 258
YUKAKO WADA AND HIROTAKA MATSUOKA

15 Analyzing Athletes’ Satisfaction toward the Organization

of a Sport Event: Development of a Questionnaire 275
ROSA MEDINA RODRÍGUEZ, OSWALDO CEBALLOS GURROLA,
MIREYA MEDINA VILLANUEVA, MARINA REYES ROBLES,
MARÍA GRETHEL RAMÍREZ SIQUEIROS, AND ROSA LÓPEZ DE D’AMICO

Index 289
Figures

2.1 Five-Factor Proposed Model. 26

2.2 Competing Models. 35
2.3 Proposed Research Model and Hypotheses. 43
2.4 Multi-group SEM Results. 46
3.1 The Disruption of the Sport Event Marketing Mix. 54
3.2 Media Devices: Communication Technological
Paradigm and Digital Disruption. 55
3.3 Features of Sports Events: From Real Events to eSport. 64
4.1 Flow Diagram. (Source: Moher et al., 2009.) 73
6.1 Networks for the LPGA. Networks Located in the
Lower Side are Adjusted by Groups (n = 368, ties = 3,708). 118
6.2 Networks for the LPGA by the Strength of Interactions (Degree). 119
6.3 Twitter Communication Networks for LPGA @LPGA and the
Impact of Distribution of Information via Angela Stanford
(@angela_stanford) and PGA (@PGA). 121
7.1 Meaning Transfer Model. (Adapted from McCracken, 1989.) 129
7.2 Consumer Purchase Decision-Making Process.
(Adapted from Solomon et al., 2019.) 137
7.3 Characteristics That Increase the Effectiveness of an
Endorser in an Online Environment. (Adapted from
Djafarova and Thompson, 2020.) 139
8.1 Facebook Insights as Displayed in the Brand Page
Administrator Console. 150
9.1 The Hypothesized Conceptual Framework. 167
9.2 Percentage of Posts and Engagement by Teams. 173
12.1 AHP Index System of Sport Facilities Service Provider Selection. 234
13.1 The Stakeholders’ Approach to Balanced Scorecard
Sports Event’s Evaluation. 253
14.1 The Structure of the Team Attributes Scale in the Strength,
Favorable, and Uniqueness Models. 266
15.1 Standardized Solution of the Structural Model of the
Satisfaction in the Organization of a Massive Sport Event. 286
Tables

2.1 Descriptive Statistics and Correlation 32

2.2 Maximum Likelihood Standardized Loadings, T-values, and
Standard Errors for the Hypothesized Structural Model (n = 200) 33
2.3 Fit Statistics and Standardized Path Coefficients for All Models
(n = 200) 36
2.4 Mediation of the Effects of Attitude, Subjective Norm, and Team
Identification on Visit Intention through the Desire to Visit
(n = 200) 38
2.5 Test for CFA Invariance of Millennials and Generation X
for Game Attending Behavior: Summary of Model Fit
and χ2-Difference-Test Statistics (N = 400) 46
4.1 Academic Variables of the Articles 75
4.2 Methodological Aspects of the Articles 77
4.3 Main Applications of the Studies 82
5.1 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Gender 100
5.2 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Age Group 101
5.3 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Affiliation 102
5.4 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Party Attending 103
5.5 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Race 104
5.6 Spectators’ Perceived Ease of Use, Perceived Usefulness,
Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Income 105
 Tables xi

5.7 Spectators’ Perceived Ease of Use, Perceived Usefulness,

Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Education 106
6.1 Numbers of Tweets, Likes, Retweets, Network Size, and Total
Ties in KLPGA and LPGA Twitter Pages 115
6.2 Categorization of Online Stakeholders in the KLPGA and
LPGA Networks 116
6.3 Eleven Groups Users in LPGA’s Network 117
6.4 30 Core Actors in the LPGA Twitter Networks Based on
Bonacich’s Beta Centrality 120
8.1 Participating Team Social Media Presence 149
9.1 Operationalization of Vivid and Interactive Post Characteristics 168
9.2 List of Categories and Variables 169
9.3 Descriptive Statistics of Explanatory Variables 169
9.4 Post Type Over Time Periods 170
9.5 Chi-Square Test Results of Time Period by Team 170
9.6 The Results of Team*Post Type*Time Period 171
9.7 Brand Attributes by Post Type (by Number) 171
9.8 Brand Attributes Over Time Periods (by Number) 172
9.9 Hierarchical Regression Analysis Results (DV: Number of Likes) 174
9.10 Hierarchical Regression Analysis Result (DV: Number of
Comments)175
9.11 Hierarchical Regression Analysis Result (DV: Number of Shares) 176
10.1 MANOVA for Event Perceptions 191
10.2 Hierarchical Regression Analysis 191
10.3 MANOVA for Event Perceptions 192
10.4 Duncan’s Post Hoc Test 192
11.1 Congruity Definitions 203
11.2 Sample Characteristics 206
11.3 Descriptive Statistics for Univariate Analysis 209
12.1 The Initial Decision Hierarchy of Sport Facilities Service
Provider Selection 225
12.2 CITC and Reliability Analysis of All 7 Intermediate Level
and 41 Lowest Level Criteria 228
12.3 CITC and Reliability Re-Analyses of the 27-Item Model
after the First Reduction 230
12.4 The Initial Analytic Hierarchy Model of Sport Facilities
Service Provider Selection 231
12.5 The Fundamental Scale of Absolute Numbers 232
12.6 The Consistency Ratio Results Matrices 234
12.7 Weight Assessment and Item Grading of the Refined Sport
Facilities Service Provider Selection Index System 235
13.1 Matrix of Key Indicators Considering the Perspectives of the
Balanced Scorecard for Major Sports Event Evaluation 254
xii Tables

14.1 Factors, Labeled Items, Definitions, and Measurement

Items for Team Attributes 265
14.2 Means, Standard Deviations, and Correlation Ranges between
Items for Strength, Favorability, and Uniqueness Models 268
14.3 Factor Loadings and Evaluation for the Goodness-of-fit of Models 269
14.4 Average Variance Extracted and the Squared Correlation
between Two Factors 270
14.5 Reliability Estimates for Models 270
15.1 Mean, Typical Deviation, Compound Reliability, and
Correlations among the Variables 279
15.2 Organization of the Rotated Factor Structure, Communalities,
Self-Values, and Explained Variance Percentages
(by Factor and Global) 280
15.3 Mean, Typical Deviation, Compound Reliability,
and Correlations among Variables 285
About This Book

The co-editors of this book have selected research papers relevant to the topi-
cal issues. In addition to this introduction chapter, this book contains a total of
14 chapters. The chapters are contributed by a total of 43 scholars representing
15 countries or territories around the world, including Australia, China, France,
Iran, Italy, Japan, Mexico, New Zealand, Serbia, South Korea, Spain, Taiwan,
United Kingdom, United States of America, and Venezuela. Co-Editors Kevin K.
Byon, Brian H. Yim, and James J. Zhang would like to take this opportunity to
thank these eminent scholars for their remarkable contributions to the comple-
tion of this book project. This book is commissioned by the WASM Executive
Board, representing this organization’s leadership and commitment to develop,
summarize, synthesize, and analyze knowledge that help enhance global sport
industry. It covers a range of key sport marketing analysis issues in both global
and local settings. The book combines scholarly output derived from diverse
inquiry protocols, such as review of literature, documentary analysis, qualitative
research, and quantitative investigations.
Contributors

Junying An is an Associate Professor Sport Management at Shanghai University

of Sport, China.
Salvador Angosto is a Researcher of the Department of Physical Activity and
Sport of the Faculty of Sports Sciences of the University of Murcia, Spain.
Kevin K. Byon is a Professor of Sport Management and Doctoral Program
Coordinator at the Indiana University, USA.
Oswaldo Ceballos is a Professor of Physical Education of the School of Sports
Organization at the Autonomous University of Nuevo Leon, Mexico.
Ching-Hung Chang is a Professor of Sport Management at the National
University of Kaohsing, Taiwan.
Wang-Chen Chen is a Master Student of Sport Management at the National
Cheng Kung University, Taiwan.
T. Christopher Greenwell is a Professor at the University of Louisville, USA.
Beth A. Cianfrone is a Professor of Sport Administration and the MS Sport
Administration Program Coordinator at Georgia State University in Atlanta,
USA.
Jason Cole is a Former Masters Student at Massey University, New Zealand.
Paul Cook is a Senior Lecturer in Sport Management at Northumbria University,
UK.
Ruth M. Crabtree has held senior positions with Higher Education in sport at
Northumbria University, UK, specializing in global teaching excellence.
Rosa López de D’Amico is the Coordinator of the Research Center
EDUFISADRED, UPEL, Maracay, Venezuela.
Yunxia Ding is an Associate Professor of College of Sports Industry and Leisure
at the Nanjing Sport Institute, China.
 Contributors xv

Jerónimo García-Fernández is an Associate Professor of Sport Management at

the Universidad de Sevilla, Spain, and CRDO of a technological university
Spin-off linked to fitness industry.
Miaad Ghasemzadeh is a Master’s Graduated at the Allameh Tabataba’i
University, Iran.
Marion E. Hambrick is an Interim Associate Dean and Associate Professor at
the University of Louisville, USA.
Christopher Hautbois is an Associate Professor at the Faculty of Sport Sciences
and Co-Head of the Master Sport, Leisure and Event Management (SLEM) at
the University of Paris-Saclay, Paris, France.
Wonseok (Eric) Jang is an Assistant Professor of Sport Communication at
Sungkyunkwan University, South Korea.
Alejandro Lara-Bocanegra is a Researcher and a Professor of the Department of
Physical Education and Sports at the University of Seville, Spain.
Shang-Chun Ma is a Professor of Sport Management at the National Cheng
Kung University, Taiwan.
Shang-Min Ma is a Professor of Sport Management and Leisure Behavior at the
National Pingtun University of Science and Technology, Taiwan.
Armin Marquez is a Lecturer of Sport Marketing and Management and
Coordinator of Program Outreach at Indiana University, USA.
Andrew Martin is a Professor of Sport Management and Physical Education at
Massey University, New Zealand.
Hirotaka Matsuoka is a Professor of Sport Marketing in the Faculty of Sport
Sciences at Waseda University, Japan.
Mireya Medina is a Professor of Sport Management of the School of Sports
Organization at the Autonomous University of Nuevo Leon, Mexico.
Rosa Medina is a Professor of Sport Management of the School of Sports
Organization at the Autonomous University of Nuevo Leon, Mexico.
Benjamin S. Mole is a Graduate in International Sport Management from
Northumbria University, UK.
Paul M. Pedersen is Professor of Sport Management at Indiana University—
Bloomington, USA.
Lidija T. Petrović is a Professor of Practice, Business & ESports at the Carolina
University, North Carolina, USA, and an Associate Professor of International
Postgraduate Educational Programmes at the University for Peace – UN
xvi Contributors

Mandated (UPEACE), Western Balkans; ECPD Belgrade, Serbia; and ECPD

Regional Institute for Development Studies, Skopje, North Macedonia.
Elena Radicchi Sport Management Lab at the University of Florence, Italy.
Maria Ramírez is a Professor and Coordinator of Research and Graduate Studies
at the State University of Sonora, Mexico.
Jason D. Reese is an Associate Professor of Marketing in the Nelson Rusche
College of Business at Stephen F. Austin State University, USA.
Marina Reyes is a Professor of Bachelor’s Degree in Sports Coaching at the State
University of Sonora, Mexico.
Jason A. Rice is an Associate Professor and Program Director of Sport
Management at North Central College in Naperville, USA.
M. Rocío Bohórquez is an Associate Professor of Social Psychology in Sport at
the Universidad de Sevilla, Spain.
Hyunseok Song is a Doctoral Student of Sport Management and Doctoral
Program Coordinator at the Indiana University, USA.
Ashleigh-Jane Thompson is a Senior Lecturer and Program Director of Sport
Management at La Trobe University, Australia.
Yukako Wada is a Lecturer of Sport Management and Marketing in the Faculty
of Sport and Health Science at the Ritsumeikan University, Japan.
Yelena Y. Wang is a PhD Student of Sport Management and Policy at the
University of Georgia, USA.
Brian H. Yim is an Associate Professor of Sport Administration at the Kent State
University, USA.
Juha Yoon is a Lecturer of Sport Management and Master’s Program Coordinator
at Indiana University—Bloomington, USA.
Patrizia Zagnoli is a Full Professor and Director of the Sport Management Lab at
the University of Florence, Italy.
James J. Zhang is a Professor of Sport Management and Director of the
International Center for Sport Management (ICSM) at the University of
Georgia, USA.
Chapter 1

Sport Marketing Analyses

An Introduction
Kevin K. Byon, Brian H. Yim,
Hyunseok Song, and James J. Zhang

Introduction
As sport leagues, teams, and events have been integral to many communities’
cultural and economic foundations, the sport industry has experienced rapid
growth and has become one of the largest industries in North America. The esti-
mated size of the sport business industry in North America has risen sharply in
recent years, from $213 billion at the end of the 1990s to approximately $540 bil-
lion in 2018 (Plunkett Research, 2018). Despite the Great Recession occurred in
2007, the sport industry has continued to grow at a rapid pace, having more than
doubled over the last two decades. Also, the sport industry has realized similar
growth trends throughout other regions of the world estimating the global sport
industry size as $1.3 trillion in 2018 (Plunkett Research, 2018; Zhang et al., 2017).
Due to the complexity of today’s sport business environment, the traditional
entrepreneurship to operate the sport organization or understanding to deal with
sport consumers is no longer sufficient to meet the needs of the market. Instead of
relying on practical experience, hunch, and extant knowledge, both sport man-
agers and marketers should engage in understanding contemporary scholarship to
allow them to bring cutting-edge information and solutions to formal professional
preparations. By doing so, the sport marketers’ analytical and problem-solving
skills improve, which further lends credence to the notion of evidence-based
sport management scholarship (Kim et al., 2021; Zhang, 2015).
This chapter is designed to introduce concepts and applications pertaining
to traditional and emerging sport marketing analyses and analytical methods to
help sport marketers better understand research process, traditional, and contem-
porary data analytics that are utilized in global sport industry. The ensuing sec-
tions address the following: (a) needs for sport marketing analyses, (b) traditional
sport marketing analytics, and (c) digital technology for sport analytics.

Needs for Sport Marketing Analyses

As indicated by the sport industry report (Plunkett Research, 2018), sport indus-
try is becoming increasingly dynamic and comprehensive as the research prob-
lems are becoming more complicated. To effectively and efficiently deal with this
DOI: 10.4324/9781003302476-1
2 K. K. Byon et al.

global phenomenon, sport marketers must be equipped with relevant analytical,

methodological, and theoretical knowledge. The sport marketing research implies
careful and systematic means of solving problems that often emerge within the
sport industry. High-quality research should involve rigorous systematic proce-
dures to (a) identify problems, (b) search and comprehend related literature, (c)
develop research questions/hypotheses, (d) determine research design and tools,
(e) collating data based on the target population, (f) analyze and interpret the
data, and (g) translate the results into actionable strategies through effective
communication (Andrews et al., 2020; Byon & Zhang, 2019).
Effective sport marketing begins and ends with the consumer; therefore, under-
standing the consumer experience is critical to strategic marketing design. Good
sport marketers seek answers to the following questions: (a) who the target sport
consumers are, (b) where they live, work, travel, and consume the sport product,
(c) where, when, and how they become exposed to the sport product, (d) how and
why they consume the sport product, (e) how they perceive the sport product,
(f) their level of satisfaction with the sport product, and (g) what their future
sport consumption behavior will be (Johnson et al., 2019; Mullin et al., 2007).
A tool for answering these important questions is marketing research: “the pro-
cess or set of processes that links the consumers, customers, and end users to the
market through information – information used to identify and define market-
ing opportunities and problems; generate, refine, and evaluate marketing actions;
monitor marketing performance; and improve understanding of marketing as a
process” (Shank & Lyberger, 2015, p. 98). In line with this definition, sport market-
ing research is the systematic process of collecting, analyzing, and reporting sport
market data to make strategic sport marketing decisions (Shank & Lyberger, 2015).
In order to understand sport consumers, sport organizations gather informa-
tion systematically and continuously. Instead of taking a reactive approach, sport
marketers proactively collect consumer information (e.g., demographic, psycho-
graphic, and product/service usage) to stay relevant in today’s highly competitive
and rapidly changing sport entertainment marketplace. A data-based marketing
(DBM) system is a comprehensive system that integrates all data files in a cen-
tralized system and enables sport marketers and organizations to retrieve target
consumer data that is useful in marketing decision-making, sales and marketing
activities, and presentations and proposals. DBM systems can be internal (e.g.,
season ticket holder list) or external (e.g., regional demographic information) via
secondary data (e.g., information collected by a government department) or pri-
mary data (e.g., fan survey) (Geurin et al., 2018; Mullin et al., 2007).

Traditional Sport Marketing Analytics

The types of data collected depend on the marketing research design, typically
one of three primary types: (a) exploratory, (b) descriptive (survey), and (c) causal
(experimental).
 Sport Marketing Analyses 3

Exploratory Design
Researchers conduct an exploratory study (e.g., phenomenology, focus groups, and
case study) when they need to increase their comprehension of a problem that is
not yet clearly defined. Using an exploratory study, researchers can finely tune a
broad problem into specific problem statements and develop potential hypothe-
ses (Shank & Lyberger, 2015). Therefore, an exploratory study is most effective
during the initial stages of researching a new problem or concept. Exploratory
research includes several techniques: secondary data analysis, pilot studies, expe-
rience interviews, and case studies. Secondary data, which are data already col-
lected but related to the research question (Shank & Lyberger, 2015), include
several types: (a) government reports and documents (e.g., Bureau of the Census
of the U.S. Department of Commerce, The Statistical Abstract of the United
States, Chambers of Commerce, Small Business Administration), (b) standard-
ized sports marketing information studies (e.g., ESPN Sports Poll, Sports Business
Research Network), (c) trade and industry associations (e.g., American Marketing
Association, European Association for Sport Management, Institute of Sport and
Recreation Management, National Sporting Goods Association, North American
Society for Sport Management, Sport Marketing Association, World Association
for Sport Management), and (d) books, journals, and periodicals (e.g., Sport
Marketing textbooks, European Sport Management Quarterly, International Journal
of Sports Marketing and Sponsorship, Journal of Sport Management, Sport Marketing
Quarterly).
An example of an exploratory study can be found in Yim et al. (2020a), who
explored millennial sport fans’ consumption traits and categorization of the gen-
eration by reviewing 22 studies published in academic journals. They found that
Markert’s (2004) generational cohort approach was the most appropriate categori-
zation for millennial sport fan studies and identified several consumption traits that
the millennial consumers showed: (a) community-driven; (b) peer pressure-
influenced; (c) emotional; (d) making their voices heard; and (e) technology-
driven. To explore the relationship spectator sport and population health, Inoue
et al. (2015) conducted a scoping study by reviewing 135 published sources.
Although the results derived from scoping review are preliminary in nature,
information gained to understand the phenomena of interest is comprehensive.
Oftentimes, the knowledge from exploratory and qualitative methods can be
served as foundation and precursor to more systematic and qualitative inquiries
such as meta-analysis.
As guides for future study and large-scale research, pilot studies are small-scale
research conducted using small sample of data (Connelly, 2008). Pilot studies are
quite commonly used in sport marketing research for following benefits: (a) it
can provide researchers ideas and insights that have not been identified before,
(b) enables preliminary testing of the hypotheses, (c) provide chance to evaluate
the usefulness of the data, (d) reduce unanticipated problems, and (e) try out
several measures and the select the measures that produce the clearest results
4 K. K. Byon et al.

for the main study. For example, Yim et al. (2020b) conducted focus group inter-
views (18 interviewees in 3 sessions) as a pilot study to explore Millennial sport
fans’ consumption behaviors and to develop a measurement scale for a large-scale
study. The focus group study enabled to examine millennial sport fans’ consump-
tion traits to compare with previously identified millennial consumption traits.
The focus group interview results confirmed four out of five millennial fans’ con-
sumption traits found in the previous literature and discovered two new traits
(rational choice and desire to be comfortable), which provided new insights that
the researchers did not identify through literature review. To examine the effect
of experiential marketing on behavioral intention in a festival with a specific
sport event, Yazıcı et al. (2017) used measurement scales developed in the United
States. The usefulness of the measures in Turkish culture was confirmed in their
pilot study where they collected a survey data to test the validity and reliability of
the scales. After confirming the usefulness of the scales though pilot study, they
used the scales in their main study.
Case study is another useful method widely used in exploratory research
(Rowley, 2002). When there is no single set of outcomes of the research ques-
tion, a (exploratory) case study is useful to inform the direction of a larger
quantitative study because even with a single case, researchers can collect deep
level data (Seaton & Schwier, 2014). Case study research method is defined as
“as an empirical inquiry that investigates a contemporary phenomenon within
its real-life context; when the boundaries between phenomenon and context
are not clearly evident; and in which multiple sources of evidence are used”
(Yin, 1984, p. 23). According to Eisenhardt (1989), case studies are well suited
for new areas of research since they can examine deep and rich qualitative
data, and for fresh perspectives on well-established areas by providing cases that
contract established ideas or theories. In the area of sport marketing, Hyatt
(2007), via the use of a well-designed case study of National Hockey League’s
Hartford Whalers, examined the effects of team relocation on the fandom of
the loyal left-behind fans. Through the case study, the author could shed light
on the experiences of left-behind fans, which was the new phenomenon in sport
consumer behavior literature.

Descriptive (Survey) Design

Researchers use a descriptive research design when the problem is more clearly
defined and the primary objective is to describe the characteristics of target
market groups (Shank & Lyberger, 2015). This approach also allows scholars to
examine the relationships or correlations among variables (Byon & Zhang, 2019).
The primary purpose of a descriptive research is to measure who, where, how
much, and how often sport marketing stakeholders (e.g., sport consumers, sport
organizations, employee, event organizers, sponsors, etc.) engage in certain con-
sumption behaviors; the most common data collection technique in descriptive
research is the survey (Shank & Lyberger, 2015).
 Sport Marketing Analyses 5

Two types of data collection methods often used in survey design are cross-
sectional study and longitudinal study. In a cross-sectional study, researchers col-
lect data at a single point in time while in a longitudinal study, researchers per-
form multiple data collections (more than three points in time) over time. There
are several advantages of cross-sectional research design that its data provide a
snapshot of the characteristics of a particular group and it is relatively inexpen-
sive and convenient research design to collect data. Although results from cross
-sectional data cannot establish cause and effect, they can reveal correlations
among variables. Cross-sectional research design is one of the most used research
designs in sport marketing research area. Some concerns about the validity of cross-
sectional studies are the common method bias (CMB) and lack of causal infer-
ence (CI) (Rindfleisch et al., 2008). Three data collection strategies are suggested
to reduce the threat of CMB and enhance CI: (a) employing multiple respond-
ents, (b) gathering multiple types of data, or (c) collecting data over multiple peri-
ods (Rindfleisch et al., 2008). To avoid the CMB stem from obtaining a predictor
an outcome from the same source at the same time point, Kim and Byon (2018)
employed a dyadic method to collect the data from two sources in two data collec-
tions when they examined the mutually beneficial relationship between employ-
ees and consumers in a fitness center. Cross-cultural studies, where data are
collected multiple times in multiple locations, in sport marketing revealed that
many human needs and psychological constructs are universal such as team iden-
tification, satisfaction, and perceived service quality (Lee & Kang, 2015; Yoshida
& Heere, 2015; Yoshida & James, 2010), which enhances the external validity.
Although abovementioned ways can minimize CMB and CI, researchers
recommend longitudinal study as a better solution to these problems (Jap &
Anderson 2004; Ostroff et al., 2002). By conducting a longitudinal study, scholars
repeatedly collect data from the same individuals at different points in time to
examine changes. According to Ployhart and Vandenberg’s (2010) definition, a
longitudinal research is “a research emphasizing the study of change and contain-
ing at minimum three repeated observations (although more than three is better)
on at least one of the substantive constructs of interest” (p. 97). The advantages
of a longitudinal research design are the possibility of measuring cause and effect
and monitoring changes of a variable (e.g., consumer decision-making) over
time (Cutler, 1979). Some disadvantages of this research design are as follows:
respondents can be conditioned, potential dropout bias, and collinearity issue
by asking same questions between times. Kunkel et al. (2016) used a longitudinal
research design to examine the development and stability of consumer’s team
brand associations and their influence on team loyalty of a new sport team in
its early development stage. They collected the survey data in three points in
time over a 19-month period from new elite-level Australian football team fans
by using newsletters including the survey links. Latent growth modeling showed
that consumers’ brand association changes over time as they get more experience
with the team. Furthermore, a cross-legged panel model revealed that brand asso-
ciations cause consumer loyalty.
6 K. K. Byon et al.

Katz et al. (2021) examined the evolution of sport fan ties and friendship ties
on Division III campuses in their longitudinal network study. A freshman orien-
tation group was recruited and their network data regarding fan ties, friendship
ties, and team identification were collected over the course of two academic years
(four points in times in four semesters) using the network questionnaire. They
used the simulation investigation for empirical network analysis for the exam-
ination of the longitudinal network data and found that sport played a signifi-
cant role in promoting friendships among new students and influenced student
success and retention. Sato et al. (2015) also used a longitudinal research design
to examine whether a distance running event can cause participants’ life satis-
faction. They used a panel data (who participated in a marathon event) to collect
data from four points in times over a five-month period. Latent growth modeling
analysis was used to analyze the longitudinal data and found that the runners’ life
satisfaction peaked immediately after the marathon event before receding.
Strictly speaking, it may not be a longitudinal analysis as there have been sev-
eral studies in sport marketing that collected data from two points of time which
still enabled to see the difference over time. Zhang et al. (2020) conducted a pre-
post study using panel data to examine the changes in relationships among event
impacts, satisfaction, and host city residents’ behavioral intentions (i.e., support
intention and word-of-mouth) before and after a major sporting event, Nanjing
Youth Olympic Games. Data were collected in two time points (three months
before and three months after the event) from the panel (i.e., residents in Nanjing
city) and researchers matched the data using email address to ensure that the
same respondents were surveyed both times. Structural equation modeling (SEM)
invariance tests were conducted to examine the pre- and post-behavioral inten-
tions and revealed that the residents’ assessment regarding sport event changed
over the six-month period. Another pre- and post-study example is Lock et al.’s
(2014) study where they collected data using two online questionnaire instru-
ments from two time points over a one-year period to examine the longitudinal
stability of, and interrelationships between, the dimensions of team identification
(i.e., affect, behavioral involvement, cognitive awareness, private evaluation, and
public evaluation) of a new Australian Rules football team. They used the Gold
Cost Suns fan database to identify the fans and collected panel data in two time
points over 12 months. Since they collected the data only in two time points, a
cross-legged panel model was used to examine the change over time. They found
that each dimension showed relative stability over time and also found that pub-
lic and private evaluation in Time 1 has a positive relationship with behavioral
involvement in Time 2.

Experimental Design
The importance of making CIs is hard to overemphasize. Experimental research
design is a method of data collection design to observe the consequences of
very specific actions or stimuli to test hypotheses under controlled conditions
 Sport Marketing Analyses 7

(Blackstone, 2014). Using experimental research, scholars can identify cause-and-

effect relationships by manipulating one or more independent variables while
holding other variables constant and measuring the impact on dependent var-
iables (Byon & Zhang, 2019; Shank & Lyberger, 2015). Experimental research
includes two broad types: (a) true experimental design and (b) quasi-experimental
design. The major difference between the two designs is random assignment.
True experimental design includes “more than one purposively created group,
common measured outcome(s), and random assignment” (Gribbons & Herman,
1997, p. 1). The individual background variables (e.g., sex, ethnicity, etc.) can-
not be purposively manipulated and their effects are eliminated though random
assignment. Therefore, the strength of true experimental research design is that it
can assure that the groups are truly comparable and that the observed differences
are due to the treatment but not due to the results of extraneous factors or pre-
existing differences of the control and experimental (or treatment) group. Through
a true experimental research design, a study procedure can be conducted rigor-
ously, which ensures the internal validity. Internal validity refers the extent to
which a study establishes a cause-and-effect relationship between a manipulation
(or treatment) and a finding (Blackstone, 2014).
The key of obtaining high internal validity is eliminating alternative explana-
tions for an outcome by minimizing confounding as much as possible. Confounding
happens when other factors potentially influence and confuse the study outcome,
which makes the cause-and-effect result untrustworthy. There are several ways
researchers can improve internal validity of a study: (a) blinding the participants
(sometimes researchers too) to make them unaware of the intervention to avoid
bias in their perceptions and behaviors; (b) manipulating an independent rather
than just observing an association without intervention; (c) random selection of
the participants to represent the population of interest; (d) random assignment
to the treatment and control group to avoid systematic bias between groups; and
(e) following rigorous study protocol for every group in the study to not to intro-
duce any other effect than the treatment (Blackstone, 2014; Campbell & Stanley,
1966). As there are many ways to establish internal validity, there are some situ-
ations that can threat internal validity. First, attrition (whether the participants
drop out of the survey) could potentially lead to biased sampling for the results.
Another threat is confounding that the third variable, other than the treatment,
influences the results. Additionally, situations such as diffusion (treatment group
interacting with control group), maturation (when study takes place over a period
of time it is possible the participants naturally changed), and repeated testing of
the measures are potential threats for internal validity (Blackstone, 2014).
However, one of the biggest disadvantages of the experimental design is that it
is artificial. The problem that the conditions of an experiment do not represent
the real-world conditions creates external validity problems. External validity
refers to how well the outcome of the experimental research can be generalized
to different measures, persons, settings, and times (Steckler & McLeroy, 2008).
External validity is threatened when experimental study does not account the
8 K. K. Byon et al.

interactions of variables in the real-world (Cuncic, 2020). Researchers ques-

tion how often the real-world social interactions can occur in the same way in
a lab (Blackstone, 2014). When random assignment is not possible due to the
constraining situations, quasi-experiments are considered to be the most useful
(Cook & Campbell, 1979). Although, due to the lack of control over random
assignment of participants to conditions, cause-and-effect results and internal
validity of the study are often questioned, quasi-experimental research design is
more common in studies conducted in real-world setting and show higher level of
external validity of the findings (Cook & Campbell, 1979). External validity can
be improved by: (a) creating psychological realism in the treatment by making
the participants believe the event of the study is a real event; (b) using statistical
method for calibration to adjust or reweigh problems related to external validity;
(c) replicating the study with different samples or in different settings; and (d)
conducing field experiments outside the laboratory (Blackstone, 2014). There are
some situations that can threat the external validity such as pre- and post-test
effects (where the cause-and-effect relationship diminishes without these tests),
sample features (particular sample is responsible for the cause-and-effect relation-
ship), selection bias (when differences between groups in a study is related to the
independent variable), and situation factors such as time of the day, location,
noise, etc. (Blackstone, 2014).
Sport marketing is one of the social science disciplines where there are many
variations due to the external factors that may be affecting the manipulated var-
iable, and it is hard to conduct a true experimental research in a laboratory due
to the interactive dynamics of sport environment, which makes the randomiza-
tion not possible. Therefore, quasi-experiments are prevalent in sport marketing
area (e.g., Asada et al., 2020; Cianfrone et al., 2011; Cianfrone & Zhang, 2006;
Jang et al., 2017; Kim et al., 2020). Jang et al. (2017) examined the effect of team
identification on spectators’ energy and happiness using an experimental research
design. They used a 2 (team identification: high vs. low) × 2 (positivity of game
outcome: winning vs. losing) between-subjects design. Participants were recruited
from Amazon Mechanical Turk (MTurk) and they were randomly assigned to
either winning or losing condition. A 6-minute highlight video of U.S. Men’s
National Soccer team’s previous games was used as the stimuli (i.e., winning and
losing outcome) in the study. They found that highly identified spectators showed
a greater level of happiness than the lowly identified spectators when their team
won the game. Both groups showed similar levels of happiness in losing situa-
tion. This experimental research established cause-and-effect between spectator-
ship and happiness while previous research findings showed inconsistent results
whether sport spectatorship positively or negatively influences fans’ happiness
(Inoue et al., 2015).
Kim et al.’s (2020) experimental research tested the mediation effect of fans’
coping strategies between the severity of spectator dysfunctional behavior (SDB)
and revisit intention and the moderating effect of self-construal (interdependence
vs. independence). Kim et al. (2020) went through a rigorous study procedure
 Sport Marketing Analyses 9

where they added two pretests for the development and validation of stimuli and
conducted repeated studies to establish external validity. Pictorial sketches were
used as a stimuli and to identify the low severity of SDB, pretest 1 was conducted
to ask participants who attended NFL games. High level of severity was physical
fighting and low level of severity was argument among spectators. Pretest 2 was
conducted to develop and validate the stimuli. In order to overcome the limita-
tion of the cross-sectional design, they conducted two studies (study 1 and study 2
to replicate and extend the study 1). Asada et al. (2020) conducted three studies
using MTurk data to analyze relative size and homogeneity influence fans’ percep-
tions and intentions to support the team. In their study 1, they used a fictional
city and a fictional soccer team in the experimental scenario which contained
high internal validity but lacked external validity. To address the limitation, they
conducted study 2, where they replicated the study with a more realistic stimulus
in which they used actual city and team names. Study 3 further improved the
external validity of the findings in the experiments.

Traditional Analytical Procedures

When a researcher is interested in comparing the differences in means of two
groups, a t-test can be used. For example, a t-test can be used when a sport mar-
keter is interested in comparing the stadium visit frequencies between fans and
non-fans. There are several types of t-tests: (a) a paired t-test, (b) a two-sample (or
independent) t-test, and (c) a one-sample t-test. A paired t-test is used when the
groups are from same population (e.g., pre and post) while a two-sample t-test is
used when comparing groups from two different populations (e.g., Cleveland fans
vs. New York fans). A one-sample t-test is performed when one group is compared
against a standard value (Bevans, 2020). When performing t-tests, researchers will
select between a one-tailed t-test and a two-tailed t-test. To find out whether or
not the two groups are different from one another, a two-tailed t-test will be used
while a one-tailed t-test is used to find out whether one group’s mean is greater
(or less) than the other group (Bevans, 2020). In sport marketing, a t-test is often
used to examine the group difference in descriptive research and in experimental
research. Researchers hypothesize that a null hypothesis that the group means
difference is zero and an alternate hypothesis that the group means difference
is different from zero, and they want to detect statistically significant difference
between groups. For example, Asada et al. (2020) used t-tests to test whether the
manipulations successfully affected study participants’ perceptions about the rela-
tive size and homogeneity. Hwang and Chung (2019) in their experimental study,
where they examined the difference in sport fans’ donation intention to college
athletics by type of communication and interactivity of communication, used a
t-test for manipulation check. They found significant difference in the mean score
of social media and traditional media. After Asada et al. (2020) and Hwang and
Chung (2019) identified the group means difference between the manipulations,
they were able to proceed to the main analysis.
10 K. K. Byon et al.

When there are more than two groups that a sport marketer wants to compare,
a t-test cannot be used. In this case, a researcher performs an analysis of variance
(ANOVA) to reveal any significant differences in a dependent variable among
three or more sport consumer groups. There are two types of ANOVA that are
commonly used, the one-way ANOVA and the two-way ANOVA. A one-way
ANOVA compares the variance in the group means considering only one inde-
pendent variable (or factor) while a two-way ANOVA compares the group means
considering two independent variables (Mackenzie, 2018). When the group
means are found to be significantly different using an ANOVA, researchers con-
duct post-hoc tests to identify which particular pairs of means are significantly
different based on the probability of at least one Type I error in a set of com-
parisons (Levine, 2014). There are many post-hoc tests and some of the most
common tests are Bonferroni procedure, Scheffe’s method, Tukey’s HSD test,
etc. (Levine, 2014). In sport marketing research, an ANOVA has been used to
compare three or more groups with one independent variable (Yim et al., 2020b)
or two independent variables (e.g., Hwang & Chung, 2019). Yim et al., (2020b)
examined sport consumption differences among fan groups from three genera-
tions, Millennials, Generation X, and Baby Boomers, using a series of one-way
ANOVA. The ANOVA results revealed that there were significant differences
in group means of sport consumption behaviors. Scheffe’s method was used for
post-hoc tests to identify group differences such as Millennials and Generation
X spent more time on sport-related social media activity participation than Baby
Boomers. In Hwang and Chung’s (2019) study, there were two independent var-
iables (media platform and interactivity), therefore they adopted a 2×2 factorial
experimental research design and used a two-way ANOVA to analyze the data.
They were able to find significant group mean difference of fans’ online donation
intention according to the media platform as well as interactivity of cause-related
marketing (CRM).

Multivariate Statistics
While ANOVA only determines group differences in a single dependent variable,
multivariate analysis of variance (MANOVA) can test for group mean differences
in two or more dependent variables. MANOVA is an omnibus test statistic and
do not tell a researcher which specific pair of groups were different from each
other, so post-hoc tests need to be tested to identify which pairs are significantly
different. In sport marketing research, scholars have used MANOVA to exam-
ine various spectator motivations (e.g., Andrew et al., 2020; Funk et al., 2009;
Pizzo et al., 2018). Andrew et al. (2020) used a MANOVA to examine gender
differences among attendance motivation (nine motivations) and found a sta-
tistically significant difference in vicarious achievement. Specifically, male fans
reported greater vicarious achievement motivation for attendance. Pizzo et al.
(2018) examined similarities and differences in spectator motives (15 motives)
between one traditional sport (Korean Professional Soccer League) and esports
 Sport Marketing Analyses 11

(FIFA Online 3 and StarCraft II) using a MANOVA. Tukey’s HSD post-hoc tests
were used to identify significant differences between context pairs. They found
similar patterns for 11 out of 15 motives and 4 different patterns out of 15 and
concluded that traditional sport and esports were similarly consumed.
Another useful parametric inferential technique is regression analysis, which
estimates the relationship between a dependent variable (i.e., outcome) and one
or more independent variables (i.e., predictor). Regression analysis is useful to
predict future trends to determine the relationship between variables and under-
stand one variable’s change when the other one(s) changes (Blokhin, 2021).
Due to its predictive power, multiple regression analysis is common in business
research (Hair et al., 2010), including sport marketing (Byon & Zhang 2019; Jin
et al., 2011). Simple regression and multiple regression are the most commonly
used types of regression analysis (Blokhin, 2021). Simple regression analyzes the
relationship between two variables (independent and dependent) in attempt to
draw a straight line based on the data by calculating the slope and interception.
However, in sport marketing research where it is rare that a dependent variable
is explained by a single independent variable, researchers use multiple regression
analysis which establishes the relationship between a dependent variable and
more than two independent variables (Blokhin, 2021).
Quarterman et al. (2013) reported that regression was the most used (23.3%)
analytic method published between 2000 and 2009 in the International Journal
of Sport Management. Because they can address particular outputs (e.g., behavio-
ral intention, satisfaction), regression models of sport consumer decision-making
are highly useful for sport marketers and organizations. An example of multiple
regression in sport marketing research is predicting sport fan satisfaction based
on five dimensions of perceived service quality inside a stadium (Theodorakis
et al., 2011). Four out of five service quality dimensions predicted the overall ser-
vice quality and the overall service quality was found to mediate the relationship
between the five service quality dimensions and fans’ satisfaction. Although mul-
tiple regression analysis is a useful analytic method, it evaluates and constructs the
relationship between variables of interval dependent variable and linear combi-
nations of interval, dichotomous or dummy independent variables. Shortcomings
of regression analysis are that it is difficult to use it to assess a complex path model
with direct and indirect effects and it cannot use latent variables with multiple
indicators to predict the dependent variables while SEM allows researchers to
overcome these shortcomings.
Factor analyses, exploratory factor analysis (EFA), and confirmatory factor
analysis (CFA) later emerged in sport marketing research where only about 3%
of the studies employed advances statistical methods such as EFA, CFA, and
SEM between 1987 and 2004 in JSM (Quarterman et al., 2006), but they gained
immediate acceptance and are now common. Byon and Zhang (2019) revealed
that until early 2000, EFA and regression, often called first-generation techniques
(Hair et al., 2010), were used by sport management researchers to identify patterns
in the data set or explore whether adding some independent variables help to
12 K. K. Byon et al.

explain more variance in the dependent variable, which helped the sport manage-
ment discipline to establish foundational knowledge and theories. Starting mid-
2000, second-generation techniques, such as CFA and SEM, have become more
frequently used in sport management research as the sport marketing research
questions were developed toward verifying the developed knowledge and proving
a priori determined factors and hypotheses (Byon & Zhang, 2019). The primary
purpose of factor analysis is to define the underlying structure among variables
(Hair et al., 2010), a process that has become popular in sport marketing research
for several reasons. First, sport marketing research involves several latent sport
consumer variables (e.g., sport fan identification, fan involvement, fan engage-
ment, spectator motivation, perceived service quality, consumer satisfaction,
and sport fan emotion). Effective measurement of these psychometric properties
requires multiple items, and factor analysis can identify how those items interre-
late. The colloquial expression “garbage in, garbage out” (Kilkenny & Robinson,
2018) applies to data and the findings they generate. To establish measurement
validity and reliability and to minimize error, sport marketing researchers use
multivariable tools, creating correlations among the variables. To manage highly
correlated variables during analysis, scholars group them into factors and, in some
cases, higher-order factors (Hair et al., 2010).
The goal of EFA is to identify the underlying relationships among measured
variables and to examine its internal reliability. It is commonly used for scale devel-
opment (e.g., Cianfrone et al., 2015; Yim et al., 2014) or when the theoretical factor
structure of the model is weak so that the researchers have no hypotheses about
the underlying factor structure, thus all the items are loaded to all factors (Byon
& Zhang, 2019). So, in EFA the number of factors is decided by examining out-
put from various factor extraction methods (e.g., a principal component analysis,
canonical factoring, alpha factoring, image factoring, and least squares) such as the
number of eigenvalues and scree plot. In an EFA procedure, the loadings are rotated
once the initial solution is obtained (Hair et al., 2010). Two basic types of rotation
are orthogonal rotation (varimax, quartamax, and equamax), which assumes the
factors are to uncorrelated to each other, and oblique rotation (oblimin, promax,
and direct quartmin) that is based on the assumption that the factors are correlated
(Hair et al., 2010). In sport marketing, Yim et al. (2014) explored and conceptual-
ized the factor structure of sport fans’ emotions that arise during attending sporting
events of their favorite team. EFA was used since the underlying factor structure was
unknown. They generated 82 initial items through literature review and reduced
the number of items via principal component analysis and Varimax rotation. Also,
items that were similarly loaded on two or more factors, items that showed low
factor loading, and items that loaded to factors that contradicted previous findings
and scales were eliminated and the number of items was reduced to 36 items within
6 factors. Eigenvalue greater than 1.0 and scree plot both recommended a six-factor
model and explained 64.8% of the variance which was adequate.
When researchers have good understanding of the theoretical factor structure,
they use CFA to verify the factor structure of a set of observed variables. CFA
 Sport Marketing Analyses 13

allows a researcher to specify which items load on which factor to assess and
confirm the hypothesized latent factor structure (Byon & Zhang, 2019; Hair et al.,
2010). Therefore, a researcher should postulate the relationship pattern a priori
based on theory, empirical research, or both to test the hypothesis. CFA typically
use maximum likelihood to estimate factor loading and it provides a fit of the
hypothesized factor structure; therefore, with CFA a researcher can assess how
well the proposed (hypothesized) model fits to the observed data. Once the model
fit is considered to be acceptable, then the results are presented and interpreted.
Hair et al. (2010) suggested to use multiple goodness of fit measures, including chi-
square statistic (χ2), normed chi-square (χ2/df), root mean square error of approxi-
mation (RMSEA), standardized root mean residual (SRMR), Tucker–Lewis index
(TLI), and comparative fit index (CFI). Many sport marketing researchers have
used CFA (e.g., Kim et al., 2020; Yim & Byon, 2020) to test whether the meas-
ures of a construct are consistent with a proposed theoretical model. Kim et al.
(2020) used AMOS 21 to conduct CFA to verify the viability of the measurement
model before conducing SEM. They specified the factor structure and relationship
based on previous empirical studies. Maximum likelihood estimation was used for
CFA and the results showed excellent model fit for the measurement model: χ2 =
450.31, df = 146, p < 0.05, CFI = 0.95, TLI = 0.94, and RMSEA = 0.08 with 90% CI
[0.07, 0.09]. After the model fit was established, they computed composite reliabil-
ity (CR) and average variance extracted (AVE) to examine construct reliability,
which turned out to be greater than the suggested cutoffs.

Partial Least Square (PLS)

According to Byon and Zhang (2019), the use of SEM has exponentially increased
in the top-ranked sport management journals (i.e., Sport Management Review,
Journal of Sport Management, and European Sport Management Quarterly). SEM
is a multivariate analysis conflating techniques of factor analysis and regression
to examine relationships among latent variables (Hair et al., 2010). As men-
tioned above, SEM is one of the most widely employed multivariate statistical
techniques in sport management and other social sciences research because of
its capability to estimate a complex research model that includes multiple inde-
pendent variables, mediators, moderators, and dependent variables. SEM can
also account for measurement error associated with its measured variables and
latent variables (i.e., disturbance). Another advantage SEM can offer is a model
comparison among multiple competing models to select the most plausible model
(MacCallum, 2003). For instance, Byon et al. (2020) examined a mediated-
moderation model of constraints, motivation, and sport consumption using U.S.
college football samples. Using SEM, the researchers could handle a complex
model involving a mediator (i.e., negotiation) and a moderator (i.e., team identifi-
cation). Using samples of Japanese and Euro-Canadian leisure participants, Kono
et al. (2020) conducted a model comparison among five theoretically plausible
competing models (i.e., the independence model, the constraint-effects-mitigation
14 K. K. Byon et al.

model, the perceived-constraints-reduction model, the negotiation-buffer model,

and the dual-channel model) to determine the optimal leisure constraint-
negotiation model. By means of SEM, a model comparison between a fully medi-
ated model and partially mediated model was reported in Byon et al.’s (2013) study.
There are two distinct approaches in assessing the relationships in a struc-
tural model. One is covariance-based SEM (i.e., CB-SEM) and the other is PLS
SEM (i.e., PLS-SEM). The former one has been the dominant approach in the
sport management research while the latter one is an emerging method in the
field. Each approach is complimentary yet distinct in terms of characteristics and
objectives. For instance, CB-SEM treats the latent variables as common factors
explicating the covariation between its measured variables. However, the latent
factors’ scores are needed in the estimation of a structural model. However, PLS-
SEM uses weighted composites of indicator variables for a construct of interest
instead of using sum scores as is the case for regression, which is an inferior esti-
mation method. Because of its use of weighted composites of variables, PLS-SEM
is often referred to as a composite-based approach as opposed to CB-SEM.
Due to the popularity of its use of CB-SEM as evidenced in Byon and Zhang’s
(2019) study, our focus is delimited to the rules of thumbs for selecting between
CB-SEM and PLS-SEM and empirical studies using PLS-SEM that are published
in sport and leisure management journals. First, when the goal of a research is
predicting the dependent variables, PLS-SEM should be used, as PLS-SEM aims
to maximize an explained variance of endogenous constructs (Hair et al., 2017),
whereas when the goal is theory testing, theory confirmation, and model compar-
ison among theoretically plausible models, then CB-SEM should be adopted, as
CB-SEM aims to minimize the difference between an implied-covariance matrix
and a sample-covariance matrix (Hair et al., 2017). For instance, Kono et al.
(2020), via the use of PLS-predict that is available in Smart PLS 3 packages, found
that the independence model demonstrated the best predictive power compared
to other alternative constraint-negotiation models.
The second situation is when a research model consists of either formatively
measured constructs or a combination of formatively and reflectively measured
constructs. In fact, CB-SEM could handle a structural model involving formative
measures. However, CB-SEM requires a complicated rule for model identification
(e.g., 2+ emitted paths rule), whereas PLS-SEM does not require such identifi-
cation rule. Therefore, PLS-SEM is considered as a preferred SEM method for
such complex models. As an example, Kim and Byon (2021) proposed the fit-
ness switching costs model (FSCM) involving a third-order reflective-formative
model to understand the antecedents of fitness consumer’s switching costs. Using
German professional soccer fans, Sarstedt et al. (2014) developed and tested a
model measuring fan satisfaction index (i.e., FANSAT) that involves 17 forma-
tively measured antecedents of fan satisfaction.
When sample size is small and/or data distribution is not normal, PLS-SEM is the
preferred estimation method over CB-SEM. For instance, Reinartz et al. (2009), in
their simulation study, found PLS-SEM, compared to its counterpart (CB-SEM),
 Sport Marketing Analyses 15

produced higher statistical power when sample size was small. According to Hair
et al. (2017), when the number of independent variables in a structural model is 5,
a minimum of 45 sample is needed to achieve a statistical power of 80% for detect-
ing 25% of variance explained in the dependent variable. However, it is impor-
tant to note that insufficient sample size is subjected to Type II error. Therefore,
researchers should not resort to the minimum sample size by solely relying on
the ability from the statistical method such as PLS-SEM. It is recommended that
researchers calculate proper sample size using programs such as G *Power regard-
less of using either CB-SEM or PLS-SEM.
The use of PLS-SEM resulted in a robust estimation with severely non-
normal data (e.g., highly skewed data), as PLS-SEM uses nonparametric approach
(Reinartz et al., 2009). Involving samples collected from distance runners, Sato
et al. (2018) tested a mediated model examining the extent to which distance
runners’ personality explains life satisfaction mediated by running involvement
by citing “PLS-SEM allows researchers to assess the validity of measurements
without strict assumptions about the distribution of the variables and sample
size required by maximum likelihood techniques” (p. 35). In brief, SEM is widely
adopted method, and it is a power method to estimate complex research models.
Exponential growth and dynamic nature in sport industry makes PLS-SEM, as
a complementary method to CB-SEM, a valuable method to address complex
research questions to advance substantive knowledge of sport management field.

Digital Technology for Sport Analytics

With advances in digital technology, sport fans can now enjoy various online
activities, but they leave valuable data behind, which generates information about
the user’s preferences, tendencies, and behavioral patterns. Like most people,
sport fans use search engines (Google, Bing, Baidu, and Yahoo!, among others)
to search for information about their interests, and they share their opinions on
social media (Instagram, Facebook, Twitter, TikTok, YouTube, and WeChat) or
online review platforms such as Google, Yelp, Amazon, and Trip Advisor. For
instance, sport fans’ internet search for big rivalry matches (e.g., Manchester derby,
El Clasico, and Boston Red Sox vs. New York Yankees) is recorded on Google
search engine, and they share their interest and supporting with Instagram pho-
tos, Twitter hastags, Facebook like, and YouTube video clips.
In the sport consumption industry, consumer interests change quickly, and it
is difficult to follow these changes in real time. However, because online data
provide timely information about consumer interests, the analytics of online
sport consumption behaviors can now reveal the constantly evolving changes in
consumer demands. To use the vast quantities of data that users generate, schol-
ars and practitioners have developed business intelligence software. Given the
emergence of the digital technology, we consider web analytics based on popular
search engines, social media analytics, and how business intelligence can apply
these data.
16 K. K. Byon et al.

Web Analytics
Web analytics, also known as web usage mining, extracts information by ana-
lyzing data generated by webpage visitors (Sharda et al., 2016). For example, a
descriptive analysis of user interest in sport is available using Google Trends (GT),
which shows that U.S. internet users over the last 12 months searched for “NFL”
approximately four times more often than they searched for “MLB”. GT also
reveals that topics of interest related to NFL include “NFL Draft”, “Washington
Football Team”, and “NFL regular season”. Web analytics is not limited to key-
word searches. An analysis of the websites of sport organizations, for example,
yields information about the online consumption behaviors of sport consumers,
such as website usability (e.g., page views, site time, downloads, click map, and
click paths), traffic sources (e.g., referral websites, search engines, direct search,
offline campaigns, and online campaigns), visitor profiles (e.g., keywords, con-
tent groupings, geography, time of day, and landing page profiles), and conversion
statistics (e.g., new visitor numbers, returning visitor numbers, online purchase,
and exit rates) (Sharda et al., 2016). One such tool is Google Analytics (GA),
which provides visitor traffic sources, behaviors, and conversion statistics by a
landing page. For instance, according to GA, it is observed that English Premier
League (EPL) fans’ conversions are diminished during packed schedules of their
supporting teams (Pasala, 2013). Based on the analytics, sport league managers
might arrange their game schedule for optimize fans’ engagement. Thus, by track-
ing consumers on their websites, organizations can analyze the number and type
of visitors, and develop market strategies such as STP strategy (segmentation,
targeting, positioning) and customer relationship marketing (CRM) perspective
accordingly.

Social Media Analytics

The rapid development of information and communication technology (ICT)
led to the emergence of social media platforms, such as Instagram, Facebook,
and Twitter, as well as online review sites, such as Google, Yelp, Amazon, and
Trip Advisor. According to one study, most internet users engage in an online
community space and the time they spend on social media is now over 2 hours
per person per day (Tankovska, 2021). In cyberspace, sport consumers share
their emotions, opinions, expectations, and experiences of sport events, ath-
letes, and sport organizations, and sponsors use social media as a collateral
communication channel with sport consumers. Although much of the data
created by social media users are unstructured and messy, new software allows
for the useful analysis of large quantities of information generated by online
users. Recently, several analytic techniques have been developed to analyze
user behaviors in social media – one, a text analytic method known as web-
scraping, reveals the preferences of sport consumers about advertising. Eddy et al.
(2021) found that followers of National Football League (NFL) and National
 Sport Marketing Analyses 17

Basketball Association (NBA) teams on Twitter engage in passive signage

tweets of sponsor activation more often than they do with the sales promo-
tional tweets of sponsor activation. Even sport fans’ engagement in social media
affects the stock price of sport clubs (i.e., Jubentus F.C., Manchester United,
and Ajax Amsterdam) influencing individual’ investing behaviors. The stock
price of sport clubs is fluctuated before and after matches, and related number
of social media conversations (Sóti et al., 2020).
Another novel technique is picture-based content analysis of the user’s expe-
riences. By using multiple machine learning applying Instagram pictures of
Austrian sport tourism, Arefieva et al. (2021) found various related keywords in
sporting activities like “urban activity”, “mountain scenery”, “lake”, and “win-
ter sporting activity” among user interests, and suggested tourist office manag-
ers to made advertising strategies related to sporting activities such as bicycle
tours or biking activities. Social media communication has also combined with
location-based information, such as geotagged tweets and Yelp’s check-ins, to
identify the tourists’ designation of choice after classifying social media users
as tourists or residents with location-based social media information. To reveal
the tourists’ spatial patterns of destination choices, multiple types of clustering
methods can be used (e.g., k-means clustering, density-based spatial clustering,
and mean-shift) (Hasnat & Hasan, 2018). Also, applying location-based social
media data, in terms of sport-event-oriented travels, Xin and MacEachren (2020)
categorized the travelers into three groups: professional athletics and colleagues,
sport industry practitioners, and sport fans. Managers in related industries (e.g.,
hotel, restaurant, and transportation) might deliver specific information and
marketing strategies toward tourists, to identify destination choices accurately.

Business Intelligence and Big Data

Due to the vast amounts of data generated by consumers, the concept of business
intelligence (BI) has become pivotal for sport-related internet analysis. While BI
is an overarching term for a firm’s information-related activities – from data min-
ing to online processing of information (Ratten & Dickson, 2020) – the goal of BI
is to find useful information from data and provide businesses with a competitive
advantage (Ratten & Dickson, 2020). Already, in the sport business industry, BI is
applied in areas such as dynamic ticket pricing, customer relationship management,
fan engagement, social media and digital marketing analytics, corporate partner-
ship evaluation, and market research (Buckstein, 2016). By tracking spectator data
(e.g., demographics, purchase history, and attendance), sport organizations can
build CRM strategies (e.g., customized marketing campaigns to target customers)
to improve their competitiveness. Because computer technology for data process-
ing has grown exponentially, it is now possible to apply large amounts of informa-
tion to BI, which consists of four areas: definition of objectives, data collection,
statistical analytics, and application to data-driven results (Ratten & Dickson,
2020). What is more, various information-related technologies (e.g., mobile
18 K. K. Byon et al.

devices, wearable devices, IoT, social media, and sensors) have boosted not only the
quantity of data but also their quality and availability. Techniques to analyze big
data are now essential for businesses to survive in the modern information-based
marketplace.
Big data has five components: volume (quantity of data), velocity (speed of
data), variety (diverse types of data), veracity (genuineness of data), and value
(usefulness of data) (Ratten & Dickson, 2020). Although big data includes a mas-
sive amount of the five Vs, it needs to be aggregated and patterned to understand
sport fan behaviors and help sport organizations to make meaningful deci-
sions. Cabrera-Sánchez and Villarejo-Ramos (2020) suggested three types of
data-analytic techniques for big data: machine learning (i.e., supervised learning,
non-supervised learning, and semi-supervised learning), data mining (i.e., deep
learning, classification, and association rules), and statistical methods (i.e., regres-
sion analysis, descriptive statistics, and inferential statistics). For example, in a
sport consumer behavior study, Twitter mentions of “World Cup” can become a
set of big data, which is then available for BI to aggregate and identify the behav-
ioral patterns of sport fans, such as basking in reflected glory (BIRG) and cutting
off reflected failure (CORF) in the sport event (Fan et al., 2020).
For example, in the 2018 FIFA World Cup semi-final match, when Croatia
defeated England, England fans’ team identification vanished on Twitter to weaken
associations. Nowadays, various types of fan-related data overflow. Although the
number of publicly accessible big data have increased, analyzing and understand-
ing the massive data are more challengeable while the data provide meaningful
clues to develop sport management strategies, and consequently sport organiza-
tions who reluctantly effort to apply sport analytics methods for digital informa-
tion might be in trouble to survive in competitive market environments.

References
Andrew, D. P. S., Pedersen, P. M., & McElvoy, C. (2020). Research methods and design in
sport management (2nd ed.). Champaign, IL: Human Kinetics.
Arefieva, V., Egger, R., & Yu, J. (2021). A machine learning approach to cluster destina-
tion image on Instagram. Tourism Management, 85, 104318.
Asada, A., Ko, Y. J., & Jang, W. E. (2020). Effects of relative size and homogeneity of
sports fan community on potential fans’ support intentions. Journal of Sport Management,
34(2), 103–119.
Bevans, R. (2020, December). An introduction to t-tests. Scribbr. Retrieved April 30, 2021,
from https://www.scribbr.com/statistics/t-test/
Blackstone, A. (2014). Principles of sociological inquiry: Qualitative and quantitative methods.
Washington, DC: Saylor Foundation.
Blokhin, A. (2021, April). Linear vs. Multiple Regression: What’s the Difference? Investopia.
Retrieved April 30, 2021, https://www.investopedia.com/ask/answers/060315/what-
difference-between-linear-regression-and-multiple-regression.asp#:~:text=Regression
%20analysis%20is%20a%20common,regressions%20with%20multiple%20
explanatory%20variables
 Sport Marketing Analyses 19

Buckstein, S. (2016). Evolution and impact of business analytics in sport. In C. K. Harrison

& S. Buckstein (Eds.), Sport business analytics using data to increase revenue and improve
operational efficiency (pp. 11–24). CRC.
Byon, K. K., Salge, C., Baker, T. A., & Jones, C. W. (2020). Team identification and
negotiation: A mediated-moderation model of constraints, motivation, and sport con-
sumption. Sport Marketing Quarterly, 29(4), 282–295.
Byon, K. K., & Zhang, J. J. (2019). Critical statistical and methodological issues in sport
management research. Measurement in Physical Education and Exercise Science, 23(4),
291–300.
Byon, K. K., Zhang, J. J., & Baker, T. A. (2013). The impact of core and peripheral service
quality on consumption levels of professional team sport spectators as mediated by per-
ceived value. European Sport Management Quarterly, 13(2), 232–263.
Cabrera-Sánchez, J. P., & Villarejo-Ramos, Á. F. (2020). Acceptance and use of big data
techniques in services companies. Journal of Retailing and Consumer Services, 52, 101888.
Campbell, D. T, & Stanley, J. C. (1966). Experimental and quasi experimental designs.
Chicago, IL: Rand McNally.
Cianfrone, B. A., & Zhang, J. J. (2006). Differential effects of television commercials,
athlete endorsements, and event sponsorships during a televised action sports event.
Journal of Sport Management, 20, 321–343.
Cianfrone, B. A., Zhang, J. J., & Ko, Y. J. (2011). Dimensions of motivation associated
with playing sport video games: Modification and extension of the sport video game
motivation scale. Sport, Business and Management, 1(2), 172–189.
Cianfrone, B. A., Zhang, J. J., Pitts, B., & Byon, K. K. (2015). Identifying key market
demand factors associated with state high school basketball tournament. Sport Marketing
Quarterly, 24, 91–104.
Connelly, L. M. (2008). Pilot Studies. MEDSURG Nursing, 17(6), 411–412.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimental: Design and analysis issues for
field settings. Dallas, TX: Houghton Mifflin.
Cuncic, A. (2020, September). Understanding internal and external validity: How these con-
cepts are applied in research. Verywell Mind. Retrieved April 29, 2021, from https://www.
verywellmind.com/internal-and-external-validity-4584479#citation-3
Cutler, F. (1979). Overview of the “Longitudinal Research Design and Analysis of Panel
Data” workshop. Advances in Consumer Research, 6(1), 31–33.
Eddy, T., Cork, B. C., Lebel, K., & Hickey, E. H. (2021). Examining engagement with sport
sponsor activations on Twitter. International Journal of Sport Communication, 14, 79–108.
Eisenhardt, K.M. (1989). Building theories from case study research. Academy of
Management Review, 14(4), 532–550.
Fan, M., Billings, A., Zhu, X., & Yu, P. (2020). Twitter-based BIRGing: Big Data analysis
of English national team fans during the 2018 FIFA World Cup. Communication &
Sport, 8, 317–345.
Funk, D. C., Filo, K., Beaton, A. A., & Prichard, M. (2009). Measuring the motives of
sport event attendance: Bridging the academic-practitioner divide to understanding
behavior. Sport Marketing Quarterly, 18, 126–138.
Geurin, A. N., Wright, C., & Zhang, J. J. (2018). Sport consumer behavior. In P. M.
Pedersen & L. Thibault (Eds.), Contemporary sport management (6th ed.) (pp. 272–291).
Champaign, IL: Human Kinetics.
Gribbons, B., & Herman, J. (1997). True and quasi-experimental designs. Practical
Assessment, Research & Evaluation, 5(14), 1–3.
20 K. K. Byon et al.

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis
(7th ed.). Upper Saddle River, NJ: Prentice Hall.
Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2017). A primer on partial least
squares structural equation modeling (PLS-SEM) (2nd ed.). Thousand Oaks, CA: Sage.
Hasnat, M. M., & Hasan, S. (2018). Identifying tourists and analyzing spatial patterns of
their destinations from location-based social media data. Transportation Research Part C:
Emerging Technologies, 96, 38–54.
Hur, Y., Lim, C. H., Won, D., & Kwon, S. (2018). Types of brand transgressions and
consumers’ moral reasoning strategies on an endorser. Sport Marketing Quarterly, 27(4),
275–286.
Hwang, G., & Chung, K. (2019). The dynamics of cause-related marketing platform
and interactivity on college sport fans’ donations. Sport, Business and Management: An
International Journal, 10(2), 227–241.
Hyatt, C. G. (2007). Who do I root for now? The impact of franchise relocation on the
loyal fans left behind: A case study of Hartford Whalers fans. Journal of Sport Behavior,
30(1), 36–56.
Inoue, Y., Berg, B.K., & Chelladurai, P. (2015). Spectator sport and population health: A
scoping study. Journal of Sport Management, 29(6), 705–725.
Jang, W., Ko, Y., Wann, D. L., & Kim, D. (2017). Does spectatorship increase happiness?
The energy perspective. Journal of Sport Management, 31(4), 333–344.
Jap, S., & Anderson, E. (2004). Challenges and advances in marketing strategy field
research. In C. Moorman & D. R. Lehmann (Eds.), Assessing marketing strategy perfor-
mance (pp. 269–292). Cambridge, MA: Marketing Science Institute.
Jin, L., Zhang, J. J., Ma, X., & Connaughton, D. P. (2011). Residents’ perceptions of
environmental impacts of the 2008 Beijing Green Olympic Games. European Sport
Management Quarterly, 11(3), 275–300.
Johnson, L. M., Chou, W. W., Mastromartino, B., & Zhang, J. J. (2019). Sport fan con-
sumption: Contemporary research and emerging trends. In C. L. Wang (Ed.), Exploring
the rise of fandom in contemporary consumer culture (2nd ed., pp. 149–170). Hershey, PA:
IGI Global.
Katz, M., Cocieru, O. C., Springer, D. L., & Dixon, M. A. (2021). Fan Ties and
Friendships: A Longitudinal Network Study of Division III Sports on Campus. Journal
of Intercollegiate Sport, 14(1), 166–188.
Kilkenny, M. F., & Robinson, K. M. (2018). Data quality: “Garbage in - garbage out”.
Health Information Management Journal, 47(3), 103–105.
Kim, E., Jeong, Y., & Zhang, J. J. (2021). Governance in a globalized sport industry. In E.
Kim & J. J. Zhang (Eds.), Sport Governance and Operations: Global Perspectives (pp. 1–8).
London, UK: Routledge.
Kim, K. A., & Byon, K. K. (2018). A mechanism of mutually beneficial relationships
between employees and consumers: A dyadic analysis of employee–consumer interac-
tion. Sport Management Review, 21(5), 582–595.
Kim, K. A., & Byon, K. K. (2021). Conceptualization of switching costs in fitness centers:
A higher-order reflective-formative model. Sport Management Review. https://doi.org/
10.1080/14413523.2021.1879565
Kim, K., Byon, K. K., & Pedersen, P. M. (2020). Coping as a mediation mechanism between
severity of spectator dysfunctional behavior and revisit intention: The moderating effects
of self-construal in sport consumption. Journal of Sport Management, 34(1), 38–52.
 Sport Marketing Analyses 21

Kono, S., Ito, E., Walker, G. J., & Gui, J. (2020). Predictive power of leisure constraint-
negotiation models within the leisure-time physical activity context: A partial least
squares structural modeling approach. Journal of Leisure Research, 51(3), 325–347.
Kunkel, T., Doyle, J. P., Funk, D. C., Du, J., & McDonald, H. (2016). The development
and change of brand associations and their influence on team loyalty over time. Journal
of Sport Management, 30(2), 117–134.
Lee, I. S., & Kang, J. H. (2015). Effects of sport event satisfaction on team identification
and revisit intent. Sport Marketing Quarterly, 24, 225–234.
Levine, D. (2014). Even you can learn statistics and analytics: An easy to understand guide to
statistics and analytics (3rd ed.). Upper Saddle River, NJ: Pearson.
Lock, D., Funk, D. C., Doyle, J. P., & McDonald, H. (2014). Examining the longitudinal
structure, stability, and dimensional interrelationships of team identification. Journal of
Sport Management, 28(2), 119–135.
MacCallum, R. C. (2003). Working with imperfect models. Multivariate Behavioral
Research, 38, 113–139.
Mackenzie, R. (2018, July). One-way vs two-way ANOVA: Differences, assump-
tions and hypotheses. Technology Networks. Retrieved April 30, 2021, from https://
www.technologynetworks.com/informatics/articles/one-way-vs-two-way-anova-
definition-differences-assumptions-and-hypotheses-306553
Markert, J. (2004). Demographics of age: Generational and cohort confusion. Journal of
Current Issues & Research in Advertising, 26(2), 11–25.
Mullin, B. J., Hardy, S., & Sutton, W. A. (2007). Sport marketing (3rd ed.). Champaign,
IL: Human Kinetics.
Ostroff, C., Kinicki, A., & Clark, M. (2002). Substantive and operational issues of
response bias across levels of analysis: An example of climate-satisfaction relationships,
Journal of Applied Psychology, 87(2), 355–68.
Pasala, P. (2013, November). Google Analytics for Spurs and Man City. GEVME. https://
web.gevme.com/why-spurs-man-city-and-everton-should-use-google-analytics/
Pizzo, A. D., Baker, B. J., Na, S., Lee, M. A., Kim, D., & Funk, D. C. (2018). eSport vs
sport: A comparison of spectator motives. Sport Marketing Quarterly, 27, 108–123.
Ployhart R. E., & Vandenberg R. J. (2010). Longitudinal research: The theory, design, and
analysis of change. Journal of Management, 36, 94–120.
Plunkett Research (2018). Sport industry statistic and market size overview, business and
industry statistics. Retrieved April 22, 2021, from https://www.plunkettresearch.com/
statistics/Industry-Statistics-Sports-Industry-Statistic-and-Market-Size-Overview/
Quarterman, J., Hwang, J., Han, K., Jackson, E. N., & Pitts, B. G. (2013). Statistical data
analysis techniques used in the International Journal of Sport Management during its first
ten years. International Journal of Sport Management, 14, 296–316.
Quarterman, J., Jackson, E. N., Kim, K., Yoo, E., Koo, G. Y., Pruegger, B., & Han, K.
(2006). Statistical data analysis techniques employed in the Journal of Sport Management.
International Journal of Sport Management, 7, 13–30.
Quarterman, J., Pitts, B. G., Jackson, E. N., Kim, K., & Kim, J. (2005). Statistical data
analysis techniques employed in Sport Marketing Quarterly: 1992 to 2004. Sport Marketing
Quarterly, 14(4), 227–238.
Ratten, V., & Dickson, G. (2020). Big data and business intelligence in sport. In V. Ratten
& T. Hayduk (Eds.), Statistical modelling and sports business analytics (pp. 25–34). New
York, NY: Routledge.
22 K. K. Byon et al.

Reinartz, W., Haenlein, M., & Henseler, J. (2009). An empirical comparison of the effi-
cacy of covariance-based and variance-based SEM. International Journal of Research in
Marketing, 26, 332–344.
Rindfleisch, A., Malter, A. J., Ganesan, S., & Moorman, C. (2008). Cross-sectional versus
longitudinal survey research: Concepts, findings, and guidelines. Journal of Marketing
Research, 45(3), 261–279.
Rowley, J. (2002). Using case studies in research. Management Research News, 25(1),
16–27.
Sarstedt, M., Ringle, C. M., Raithel, S., & Gudergan, S. P. (2014). In pursuit of under-
standing what drives fan satisfaction. Journal of Leisure Research, 46(4), 419–447.
Sato, M., Jordan, J. S., & Funk, D. C. (2015). Distance running events and life satisfac-
tion: A longitudinal study. Journal of Sport Management, 29(4), 347–361.
Sato, M., Jordan, J. S., Funk, D. C., & Sachs, M. L. (2018). Running involvement and life
satisfaction: The role of personality. Journal of Leisure Research, 49(1), 28–45.
Seaton, J. X., & Schwier, R. A. (2014). An exploratory case study of online instructors:
Factors associated with instructor engagement. International Journal of E-Learning and
Distance Education, 29(1), 2–16
Shank, M. D., & Lyberger, M. R. (2015). Sports marketing: A strategic perspective. London,
UK: Routledge.
Sharda, R., Delen, D., & Turban, E. (2016). Business intelligence, analytics, and data science:
A managerial perspective (4th ed.). New York, NY: Pearson.
Sóti, A., Ancarani, A., Stéger, J., & Vattay, G. (2020). Influence of Twitter activity on the
stock price of soccer clubs. Social Network Analysis and Mining, 10(1), 1–12.
Steckler, A., & McLeroy, K. R. (2008). The importance of external validity. American
Journal of Public Health, 98(1), 9–10.
Tankovska, H. (2021). Daily time spent on social networking by internet users worldwide
from 2012 to 2020, Statista. https://www.statista.com/statistics/433871/daily-social-
media-usageworldwide/#:~:text=As%20of%202019%20and%202020,minutes%20
in%20the%20previous%20year
Theodorakis, N., Alexandris, K., & Ko, Y. J. (2011). A service quality framework in the
context of professional football in Greece. International Journal of Sports Marketing and
Sponsorship, 12(4), 57–71.
Xin, Y., & MacEachren, A. M. (2020). Characterizing traveling fans: A workflow for
event-oriented travel pattern analysis using Twitter data. International Journal of
Geographical Information Science, 34(12), 2497–2516.
Yazıcı, T., Koçak, S., & Altunsöz, I. H. (2017). Examining the effect of experiential mar-
keting on behavioral intentions in a festival with a specific sport event. European Sport
Management Quarterly, 17(2), 171–192.
Yim, B. H., & Byon, K. K. (2020). Critical factors in the sport consumption decision
making process of millennial sport fans: A revised model of goal-directed behavior.
International Journal of Sports Marketing and Sponsorship, 21(3), 427–447.
Yim, B. H., Byon, K. K., & Baker, T. A. (2014). Conceptualization and empirical verifi-
cation of the sport fan emotional experience. International Journal of Asian Society for
Physical Education, Sport and Dance, 12, 19–39.
Yim, B. H., Byon, K. K., Baker, T. A., & Zhang, J. J. (2020b). Identifying critical fac-
tors in sport consumption decision making of millennial sport fans: Mixed-methods
approach. European Sport Management Quarterly. https://doi.org/10.1080/16184742.
2020.1755713
 Sport Marketing Analyses 23

Yim, B. H., Byon, K. K., Baker, T. A., & Zhang, J. J. (2020a). Generational market seg-
mentation of millennial sport fans: Application of the generational cohort approach. In
B. G. Pitts & J. J. Zhang (Eds.), Global sport management studies: Perspectives on the USA
sport business industry (pp. 154–175). London, UK: Routledge
Yin, R. K. (1984). Case study research: Design and methods. Thousand Oaks, CA: Sage.
Yoshida, M., & Heere, B. (2015). Sport marketing in Asia: Exploring trends and issues in
the 21st century. Sport Marketing Quarterly, 24(4), 207–213.
Yoshida, M., & James, J. D. (2010). Customer satisfaction with game and service experi-
ences: Antecedents and consequences. Journal of Sport Management, 24, 338–361.
Zhang, J. J. (2015). What to study? That is a question: A conscious thought analysis.
Journal of Sport Management, 29(1), 1–10.
Zhang, J., Byon, K. K., Xu, K., & Huang, H. (2020). Event impacts associated with res-
idents’ satisfaction and behavioral intentions: A pre-post study of the Nanjing Youth
Olympic Games. International Journal of Sports Marketing and Sponsorship, 21(3),
487–511.
Zhang, J. J., Huang, R. H., & Wang, J. J. (2017). Introduction. In J. J. Zhang, R. H.
Huang, & J. Nauright (Eds.), Sport business in leading economies (pp. 1–26). New York,
NY: Emerald.
Chapter 2

Application of Structural Equation

Model in Sport Marketing Analyses
Kevin K. Byon, Brian H. Yim,
Junying An, and James J. Zhang

Introduction
Structural equation modeling (SEM) has become one of the most widely
adopted advanced statistical methods in social science research. The Sport
management field is not immune to this phenomenon. Two reasons have made
this possible: (a) An SEM model can accommodate more complicated research
problems that involve several independent variables, mediators, moderators,
and more than one dependent variables, and (b) increased availability of
user-friendly statistical software packages (e.g., model-fitting software), such
as IBM SPSS Amos, Mplus, LISREL, EQS, and SAS. Despite the advantage
and accessibility, scholars in sport management have still struggled to better
understand and appropriately apply SEM to their research. To make matters
worse, it has often been reported that SEM is one of the most widely abused
statistical methods (Byrne, 2012; Kline, 2011). This phenomenon is often
observed in sport management research by using inadequate procedures, and
reporting insufficient results makes scholars unable to reproduce the results.
It is important to note that SEM is not the path to a successful publication
or panacea but a valuable tool to confirm or disconfirm theories that help
researchers generate new knowledge or verify existing knowledge. However,
to the best of the authors’ knowledge, there is no single manuscript or mon-
ograph that details SEM in terms of definition, key concepts, recommended
procedures, and advanced SEM techniques existing in the sport management
literature.
The purpose of the current study is to introduce concepts and applica-
tion procedures related to SEM to the sport management community. We
expect that this chapter can be used as a reference to help sport management
scholars better understand and appropriately apply SEM as it helps to answer
important research questions. The following sections address the following:
(a) the definition of SEM, (b) recommended step-by-step SEM procedures,
(c) advanced SEM procedures, and (d) illustrated examples showing how to
conduct SEM using sample data.

DOI: 10.4324/9781003302476-2
 Application of Structural Equation Model 25

What Is SEM?
SEM is defined as “a technique that we use to analyze ‘causal’ influence relating
latent variables along with their attendant measured indicator variables to one
another” (Meyers et al., 2016, p. 583). SEM is a theory-driven data analytical tech-
nique that evaluates a priori specified hypotheses. Specifically, a theory-driven
analysis (confirmatory) sets SEM apart from an exploratory analysis that is data-
driven. An SEM model is comprised of at least two measurement models that
have linked to each other. In other words, two or more constructs, both of which
are defined by sets of observed (measured or indicator) variables, are theorized to
examine a cause and effect by testing the relationship between the two or more
constructs. Another characteristic that makes SEM valuable is the ability to cor-
rect for measurement error, of which other multivariate statistical procedures are
not capable. Even regression analysis assumes that there was no error in the inde-
pendent variables. When the error is substantial in the explanatory variables,
the results that were not taken into consideration will immensely mislead read-
ers. SEM can incorporate error(s) that exist not only in the observed variables
(i.e., measurement error) but also those are associated with latent variables (i.e.,
disturbance). Lastly, the SEM model can be theorized to incorporate multiple
independent mediators, moderators, and dependent variables. SEM procedures
can simultaneously test a complicated model involving numerous latent variables,
reducing Type I error compared to other multivariable statistical procedures such
as regression analysis.
Given these useful features, SEM has become a widespread choice for social sci-
entists whose design is non-experimental research (e.g., Byon et al., 2020; Zhang,
Byon, Xu, et al., 2020). The following sections address recommended step-by-step
SEM procedures that include a detailed explanation of some key concepts pertain-
ing to SEM. In explaining the concepts and procedures, we provide an illustrative
example of how to use and report SEM for research articles using empirical data.

General Procedures
Kline (2011) suggested six steps for SEM analysis: (a) specify the model; (b) evaluate
model identification (if the model is not identified, go back to step 1); (c) select the
measures to report (operationalize the constructs) and collect, prepare, and screen
the data; (d) estimate the model; (e) respecify the model if necessary based on the-
oretical and sound empirical reasons; and (f) report the results. Other scholars have
suggested similar procedures (e.g., Hair et al., 2010; Schumacker & Lomax, 2010).

Model Specification
The specification is defined as a “representation of your hypotheses in the form
of a structural equation model” (Kline, 2011, p. 92). Researchers agree on the
26 K. K. Byon et al.

importance of model specification and suggest that it is the most challenging

step in SEM analysis (Hair et al., 2010; Kline, 2011). Kline (2011) argued that this
step is essential because it confirms that the model is correct and ready for the
subsequent steps. This argument indicates that SEM represents a theory-based
data analytical method involving a priori hypothesis examining the relationship
among latent variables. Because of the theory-driven aspect, many models that
deem theoretically meaningful can be generated, which are termed “equivalent
models”, which can also be used as a counter-model against the originally pro-
posed model. Thus, researchers must provide a rationale based on solid theory
and empirical evidence in the model specification stage.
Researchers assign relationships between one latent construct and another
based on the proposed theoretical model during model specification. A model
specification often represents model parameters in the form of a diagram. The
observed variables are specified as either squares or rectangles, and the latent
variables are represented by circles or ellipses. The hypothesized directional effect
of one variable on another variable is represented via a single arrowhead pointed
from exogenous (independent) variable to endogenous (mediating or dependent)
variable (see Figure 2.1). In addition, a curved line with two arrowheads between
two exogenous constructs is used to show covariance in an unstandardized solu-
tion or correlations in a standardized solution.
Drawing a single arrowhead from one variable to another is not a simple task
as it may sound. Strong theory must support for a theoretical relationship because
it presumes a causal effect (Kline, 2011). For instance, Yim and Byon (2020) pro-
posed the sport fan model of goal-directed behavior (SFMGB), a structural model,
to predict millennial sport fan consumption behavior. In their hypothesis devel-
opment, they provided detailed theoretical justifications to support the causal
relationships specified among the latent variables. In SEM, a model specification
derived from theory and empirical evidence is a “must-have” procedure because

Figure 2.1 Five-Factor Proposed Model.

 Application of Structural Equation Model 27

results from SEM analysis serve as either confirm or disconfirm a causal structural
relationship among constructs of interest. Establishing a causal relationship is an
imperative yet difficult procedure in science. Several procedures can be adopted
to demonstrate a causal relationship. For instance, James et al. (1982) proposed
a list of ten conditions. The first seven conditions establish a firm theoretical
ground before testing the model, and the last three empirically test the causality
assumptions using sample data.
The first condition is a formal statement of theory in terms of a structural model.
Establishing a relationship between two or more variables requires a rationale
based on the theory being used in the model (study). Moreover, the researchers
should be able to display a causal relationship between variables graphically. The
second condition is the theoretical rationale of causal hypotheses. When two or
more variables have covariance, they do not automatically have a causal relation-
ship. One analogy is the covariant relationship between the moon and the sun.
When the moon goes down, the sun comes up, but we would be mistaken to think
that the moon’s disappearance causes the sun to rise. No theoretical rationale
exists to support this causal relationship. The third condition is a specification of
causal order. One should specify the sequence of the variables within the causal
model. For example, suppose we wish to examine the causal relationship between
satisfaction and repurchase intention. In that case, we must provide theoretical
justifications for why satisfaction is a cause of repurchase intention.
The fourth condition is a specification of the causal direction. Directionality
between variables can be one-way or two-way. Uni-directionality means that the
causality between two variables flows in one direction only; bi-directionality means
that the causal relationship is reciprocal, that each variable influences the other
simultaneously. Failure to specify “a priori” in confirmatory analyses (e.g., confirm-
atory factor analysis [CFA] and SEM) will lead to an invalid post-hoc study. The
fifth condition is self-containment. All possible reasons (i.e., variables) should be
considered and included in the theoretical model to draw causality. One example
might be the relationship between the height of a child and nutrition. We know a
positive relationship exists between nutrition and the height of a child. However,
other causal factors might exist along the pathway from nutrition to height (e.g.,
DNA and environment). Considering these other factors can strengthen the argu-
ment made for the theory (i.e., the relationship between height and nutrition). The
sixth condition is the specification of boundaries, which is how far we can generalize
beyond the context in which the model was tested. In one context, the model might
work very well, but the causality might not exist in another context. For example,
certain organizational cultures in Asian countries (e.g., “Gung Ho”) might influ-
ence how willing the company’s employees are to purchase its products. However,
this phenomenon might not exist in the United States or European countries. The
seventh condition is the stability of the model. If the model is stable, it will show
invariance between repeated measures and yield similar results in different contexts.
The last three conditions deal with the empirical testing of causality. The
eighth condition, operationalization of variables, refers to various questions related
28 K. K. Byon et al.

to measurement, including (a) how do we measure the variables? (b) which items
do we use to measure them? And (c) at what time intervals should we meas-
ure them? If the first seven conditions are met, this step will not typically raise
any problems. The ninth condition is empirical confirmation of the model proposed
model. In this stage, we test the model using the sample data collected in the
eighth condition. Then, researchers should confirm all of the hypothesized paths,
and we should specify all the errors that stem from the non-measured variables in
the model. Finally, the tenth condition is empirical disconfirmation of the omitted
direct paths in the model. Once we have tested and confirmed all the paths in the
model, we need to test the model to disconfirm all of the direct paths omitted
from the model. By doing so, we confirm that no causal relationships are missing
from the tested model.
Moving through these rigorous stages to infer causal relationships can help
researchers avoid misspecification issues. When the true model that generated
the data is not consistent with the specified theoretical model, the theoretical
model is said to be “misspecified” (Schumacker & Lomax, 2010). Misspecification
can be easily detected by examining model fit indices. The most common reasons
for model misspecification are errors of omission and/or inclusion of variables or
parameters that do not fit the true model (Schumacker & Lomax, 2010). In other
words, the proposed model was not generated based on sound theory and strong
empirical evidence. Following the procedure suggested by James et al. (1982) can
help researchers develop a good causal model and avoid model misspecification.

Model Identification
In SEM, researchers must resolve identification issues before data collection and
estimation of parameters of a proposed model. Model identification presents the
following question: “Can a unique set of parameter estimates be found through
the sample covariance matrix S and the theoretically proposed covariance matrix
Σ?” The type of model identification depends on the amount of information avail-
able in the sample variance-covariance matrix S (i.e., known elements) to esti-
mate the unique parameters of the model (i.e., unknown elements). Structural
equation models must be over-identified (at least just-identified) to be meaning-
fully analyzed. Of the three types outlined below, the first two indicate that the
model is identified:

1 A model is just-identified if all the parameters are uniquely determined based

on sufficient information in the matrix S (i.e., has the same number of free
parameters as observations).
2 A model is over-identified when one or more parameters can be estimated in
more than one way. The amount of information in the matrix S is more than
sufficient (i.e., it has fewer free parameters than observations).
3 A model is under-identified (or not identified) when none of its parameters
can be uniquely estimated due to a lack of information in the matrix S.
 Application of Structural Equation Model 29

When a model is under-identified, the degrees of freedom for the model will be
negative, and the parameter estimates would be untrustworthy. Several condi-
tions must be met to establish model identification. Among those conditions,
the order condition is the easiest one to identify. In other words, the number of
free parameters estimated in a theoretical model must be less than (or equal to)
the number of distinct values in the matrix S. The formula to calculate a distinct
value is p(p + 1)/2, where p is the number of observed variables. However, the
order condition, while necessary, is not sufficient. The rank condition is a suffi-
cient condition but requires an algebraic determination of parameter estimation
(see Bollen, 1989, pp. 98–103; Jöreskog & Sörbom, 1996).

Model Estimation and Model Fit Indices

The derivation of parameter estimates is based on statistical estimation meth-
ods. The most common estimation method used in the sport management field
is the maximum likelihood (ML) method (e.g., Byon et al., 2013; Lee et al., 2016;
Zhang & Byon, 2017). ML is based on normal theory in multilevel structural
equation models (Ryu, 2011) based on two assumptions: (a) the population dis-
tribution of the endogenous variables is normal, and (b) the data are multivar-
iate normal (Kline, 2011). Because the default estimation method of most SEM
software is ML, researchers conduct data normality checks to ensure the data
meet the standard of multivariate normality. Checking normality at the uni-
variate and multivariate level can be done in several ways: (a) scatter plot and
Q-Q plot examination, (b) skewness and kurtosis measures, and (c) Mardia’s
normalized multivariate skewness and kurtosis test. When these tests are suc-
cessful, researchers can use ML to estimate the parameters. Because ML is the
default method, sport management researchers do not typically report its usage
in their data analysis methods. Still, Schumacker and Lomax (2010) recom-
mended reporting estimation method types used in the results section. When
the normality assumptions are shown to be untrue, and ML cannot be used,
researchers should use alternative methods of estimation, such as weighted least
squares (WLS), generalized least squares (GLS), and asymptotically distribution
free (ADF) (Hair et al., 2010). In these cases, the researcher must provide theo-
retical justification for the chosen estimation method. For instance, Magnusen
et al. (2012) found normality violations in their data at both the univariate
and multivariate levels. Therefore, they justified their use of the Satorra–
Bentler (S-B x 2 ) scaling method (Satorra & Bentler, 2001). S-B x 2 considers
the non-normality condition and corrects the maximum likelihood chi-square
variate and the standard errors of parameter estimates (Finney & Distefano,
2006). This adjusted estimation is Satorra–Bentler’s maximum likelihood mean
adjusted (MLR) estimation method, found in the Mplus software application.
Zhang et al. (2006), in their development of a scale for program facilitators to
measure the effectiveness of after-school achievement programs, utilized the
WLS estimation method to conduct the CFA.
30 K. K. Byon et al.

For SEM, assessing model fit is important, as the goal is to minimize the dis-
crepancy between an implied-covariance matrix and a sample-covariance matrix.
When the model fit is not satisfactory, the structural model does not represent the
collected data well, and a modification to the model is necessary based on strong
theoretical justifications and another set of data. Model fit indices are controver-
sial for several reasons. First, only the chi-square test is a true statistical test of
model fit; however, as sample size increases above 200 (see Hu & Bentler, 1999),
the likelihood of type II error (i.e., false rejection) increases. Given that SEM is an
asymptotic (i.e., large sample) technique, this pattern is problematic. Therefore,
other model fit indices have been developed to address these limitations. For
instance, Tanaka (1993) identified six dimensions of model fit: (a) absolute
vs. relative (incremental), (b) simple vs. complex, (c) normed vs. non-normed,
(d) population vs. sample, (e) estimation method independent vs. estimation
method dependent, and (f) sample-size dependent vs. sample-size independent.
Kline (2011) suggested three types: absolute, relative, and predictive fit indices.
Sport management researchers typically report the following fit indices (e.g., Byon
et al., 2013; Magnusen et al., 2012; Trail et al., 2012; Wright et al., 2017; Yoshida
& Gordon, 2012; Zhang, Byon, Tsuji, et al., 2020).

Chi-square
This index should always be reported. It is an exact test of whether the model
fits the data. Because it is a “badness of fit” index (i.e., a small value is better),
researchers hope to find a non-significant chi-square. A problem with the chi-
square index is that it is sensitive to the sample size, and most of the time, the
results are significant. Therefore, alternative fit indices are needed to be exam-
ined. We discuss the other fit indices widely used in SEM studies.

Normed Chi-square
This index is the chi-square divided by the degrees of freedom. In an attempt to
mitigate the impact of the sample size issue on the chi-square index, research-
ers have subscribed to the idea that their problems will be fixed if they use the
normed chi-square. Bollen (1989) recommended a maximum ratio cut-off of 2.0 or
3.0; however, he mentioned that a cut-off of as high as 5.0 is adequate.

Standardized Root Mean Square Residual (SRMR)

This index is calculated as the difference between the observed and reproduced
correlation matrices. It is an absolute fit index, where the lower results are better,
and zero indicates a perfect fit. Hu and Bentler (1998) recommended a cut-off
of 0.05 for good fit and 0.08 for acceptable fit. However, the mean value of 0.08
could indicate several correlation residuals greater than 0.08. Kline (2011) recom-
mended supplementing this index with a visual inspection of the residual correla-
tion matrix. Because this measure is based on population, it is not dependent on
 Application of Structural Equation Model 31

the estimation method. SRMR is sensitive to variance/covariance differences and

should be used with other measures such as CFI (Hu & Bentler, 1998).

Root Mean Square Error of Approximation (RMSEA)

This population-based measure is an absolute function that corrects for sample size
(N − 1 in the denominator) and favors simpler models (i.e., uses degrees of free-
dom). It is also based on non-centrality, meaning that it can have a confidence
interval attached to it. Lower values are better, but a common mistake is to think
zero indicates a perfect fit. In this case, zero indicates that the chi-square is greater
than the degrees of freedom. Hu and Bentler (1998) recommended that zero be in
the confidence interval for a good fit. Hu and Bentler (1999) recommended a cut-off
of ≤ 0.06. Any greater value indicates a bad fit. This measure is estimation method
free; if researchers establish a chi-squared value, they can use the formula. This
measure is useful in comparing nested models (e.g., Vandenberg & Grelle, 2009)
and can also be used to evaluate the fit of the segregated structural component of
a structural model in nested models. Many sport management scholars report both
SRMR and RMSEA in their SEM results section (e.g., Byon et al., 2013; Magnusen
et al., 2012; Trail et al., 2012), but sometimes, researchers only report RMSEA (e.g.,
Lee et al., 2016; Wright et al., 2017; Yoshida & Gordon, 2012).

Comparative Fit Index (CFI)

CFI is a normed, relative, population-based measure that favors complex models.
Because it is most sensitive to factor loadings, it nicely balances the properties of
SRMR. CFI is not very sensitive to sample size. According to Hu and Bentler (1998),
a score of ≥ 0.95 is a cut-off value for the goodness of fit. Hair et al. (2010) relaxed
the cut-off value to 0.90 or above. CFI is usually reported in the SEM results of sport
management studies (e.g., Byon et al., 2013; Lee et al., 2016; Magnusen et al., 2012;
Trail et al., 2012; Wright et al., 2017; Yoshida & Gordon, 2012).

Tucker–Lewis Index (TLI)

TLI is a non-normed, relative, sample-based measure that is not very sensitive to
sample size. It is a comparison of the normed chi-square values for the null and
specified model. The value can range from 0 to 1, and a value close to 1 is con-
sidered a good model. Similar to CFI, a value equals .90 or above is considered a
good fit. In sport management studies, TLI is not as often reported (e.g., Wright
et al., 2017; Zhang, Byon, Xu, et al., 2020).

Akaike Information Criterion (AIC)

AIC is a predictive measure, and while it has two different formulas used to cal-
culate it, lower numbers mean that the solution will be reproducible (Kline, 2011).
It is sample-based and insensitive to the calculation method. Comparing models
32 K. K. Byon et al.

using AIC requires that the same sample be used. Compared to BIC, AIC does
not perform well in simulations because AIC does not assume a perfect model.
Instead, it assumes a best-fitting model among alternatives. From a conceptual
viewpoint, AIC is more accurate for sport management studies (e.g., Zhang et al.,
2006). In our attempts to represent processes that involve human behavior, we
will never design a perfect model. This fit index is often used to compare non-
nested models.

Bayesian Information Criterion (BIC)

Although AIC is better than BIC, reporting both indices is common. BIC is
population-based, assumes that a true model exists, and is not estimation method
sensitive.

Selecting Measures to Report

Various SEM theorists have proposed guidelines for reporting SEM results (e.g.,
Hoyle & Panter, 1995; Maxwell & Cole, 1995; Raykov et al., 1991; Schumacker
& Lomax, 2010). Generally, the first item on the checklist is to review findings
that support the theoretical model. Most SEM studies in the field of sport man-
agement support model specification with theoretical reasoning. The second
item is to indicate the type and version of SEM software used. For example,
Magnusen et al. (2012) reported that they used M-plus 6.1. The third item is
to indicate the types of SEM model analysis used. The fourth item includes
sample size and the correlation matrix, mean values, and standard deviations of
the variables. With this information, analyzed models can be reproduced. See
Table 2.1 for a sample table that includes correlation matrix, sample size, mean,
and standard deviation.

Table 2.1 Descriptive Statistics and Correlation

Correlation Matrix (n = 200)

Construct Mean SD 1 2 3 4 5

1 Attitude toward attending 6.21 .99 1.00

a sport event
2 Subjective norm for 5.59 1.13 .44* 1.00
attending a sport event
3 Team identification 6.25 .97 .48* .47* 1.00
4 Desire to attend a sport 5.81 1.05 .71* .54* .49* 1.00
event
5 Intention to attend 5.41 1.70 .54* .43* .38* .61* 1.00
a sport event

Note: Two-tailed; SD: standard deviation.

* p < .01.
 Application of Structural Equation Model 33

Table 2.2 M
 aximum Likelihood Standardized Loadings, T-values, and Standard Errors
for the Hypothesized Structural Model (n = 200)

Path coefficients B T SE

Attitude toward attending a sport event → Desire 0.512** 8.794 0.058

Subjective norm for attending a sport event → Desire 0.318** 5.052 0.063
Team identification → Desire 0.113 1.805 0.063
Team identification → Intention 0.177* 2.473 0.072
Desire → Intention 0.518** 8.093 0.064

Note:
* p < .05.
** p < .001.

The fifth item is to include the proposed model in the form of a diagram. The
sixth item that should be reported is the fit indices used and the reasons they
were chosen. Sample size, power, and effect size reports are also suggested. Next,
validity (e.g., convergent validity, discriminant validity) and reliability (e.g.,
Cronbach’s alpha, construct reliability, average variance extracted) tests results
should be reported. Lastly, the path coefficient table should be reported to give
readers a better understanding of the overall SEM results. Many SEM studies
include a path coefficient table (e.g., Byon et al., 2013; Magnusen et al., 2012) (see
Table 2.2).

Model Modification
Achieving an acceptable fit index is necessary to test hypotheses, but the chance
is high that a theoretical model will not have an acceptable fit initially (Hair
et al., 2010; Schumacker & Lomax, 2010). Therefore, a model modification is
typically necessary. Most model modifications are conducted at the measurement
model level (i.e., CFA model) because it is a primary source of misspecification
errors (Schumacker & Lomax, 2010) and the foundation of the structural model.
Researchers can conduct model modification in several ways, but all modification
decisions must be based on theoretical reasoning, not by the empirical reasoning
provided by SEM software (Hair et al., 2010; Kline, 2011). When model fit indices
are unacceptable, researchers can use chi-square difference test results to compare
the measurement and structural models and determine whether any additional
paths might improve the model fit (Hair et al., 2010). The next step is to exam-
ine the standardized residuals and modification indices. Eliminating items hav-
ing a large standardized residual and modification index can improve the model
(Hair et al., 2010). After modification, to cross-validate, the model, running the
model with another set of data is suggested, and that additional estimation pro-
cedure can demonstrate that the model is revised based on theory, not by chance
(Schumacker & Lomax, 2010).
34 K. K. Byon et al.

Model Comparison
Good model fit does not guarantee that the proposed theoretical model is the
best explains the observed sample covariance matrix (Hair et al., 2010). It is just a
theoretically plausible model. Equivalent models are always possible (Schumacker
& Lomax, 2010); therefore, it is highly recommended that researchers create rival
models based on theory and compare them to the initially hypothesized model
(see Byon et al., 2013; Byon et al., 2018). Doing so is a proper way to show that the
proposed model is one of the plausible models that explain the phenomenon of
interest. Researchers can compare rival models via model fit indices, chi-square
difference test, and R-square values (for nested models) and via model fit indices
of expected cross-validation index (ECVI) and AIC (for non-nested models) to
determine the best plausible model (Brown, 2015). Figure 2.2 shows an example
of SEM model comparison.
As indicated in Table 2.3, all the model fit indices for Model A were better
than the other models. The chi-square difference tests also showed that Model
A was better than the others. For example, the chi-square difference between
Model A and Model B was 6.14, with a 1 degrees of freedom difference resulting
in a p-value less than <.05. This result shows that the proposed model had one
more parameter but was significantly better. On the other hand, the chi-square
difference between Model A and Model C was 3.21, which is not significant at a
p-value of <.05 level. However, all the Model A showed better model fit indices
and higher R2 value. Therefore, we select Model A to be an optimal model among
the rival models.
Thus far, we discussed the definition of SEM, the advantages of using SEM over
other estimation procedures, such as regression, and suggested SEM procedures
that can be adopted by sport management scholars. Recently, advanced SEM
procedures have been employed to estimate a highly complex research model,
including mediators (e.g., Magnusen et al., 2012; Yim & Byon, 2020) or/and mod-
erators (e.g., Yoshida & Gordon, 2012; Yim et al., 2018; Zhang, Byon, Xu, et al.,
2020). In the following section, we will discuss advanced SEM procedures that are
useful to help answer more complicated research questions.

Advanced SEM Procedures

Mediation
A mediating effect is assumed when a third variable intervenes between related
constructs and helps explain why a relationship exists (Hair et al., 2010). Sport
management researchers theorized models by incorporating many mediating var-
iables (e.g., Magnusen et al., 2012; Trail et al., 2012; Yim & Byon, 2020; Yim et al.,
2018). Several tests can assess mediating effects in SEM analysis. Of those, Baron
and Kenny’s (1986) approach has been a popular choice, which follows the four
steps. Their causal steps approach linearly specifies the relationships among the
 Application of Structural Equation Model 35

Figure 2.2 Competing Models.

Note: Model A is the originally proposed model, and Models B and C are rival models (alternative
models). In order to compare them, we considered fit indices (i.e., χ , normed fit, CFI, TLI,
2

and RMSEA) and R-square values (see Table 2.3).

36 K. K. Byon et al.

Table 2.3 Fit Statistics and Standardized Path Coefficients for All Models (n = 200)

ATT → SN → TID → TID → Des →

Fit Index Des Des Des Des Int

χ2 df χ2/df CFI TLI RMSEA R2 β β β β β

Model A 181.71 69 2.63 0.96 0.95 .090 0.40 0.51*** 0.31*** 0.11 0.17* 0.51***
Model B 187.57 70 2.67 0.96 0.94 .092 0.38 0.51*** 0.32*** 0.12* NA 0.61***
Model C 184.92 70 2.64 0.96 0.95 .091 0.39 0.56*** 0.35*** NA 0.18** 0.52***

Note: ATT = Attitude toward attending sport event; SN = Subjective norm for attending sport event;
TID = Team identification; Des = Desire to visit; Int = Intention to visit.
* p < .05.
** p < .01.
*** p < .001

independent, dependent, and mediator variables in four separate regression mod-

els. These four steps are listed below:

1 Demonstrate that the independent variable (X) is significantly related to the

dependent variable (Y). This is denoted as the c path.
2 Demonstrate that the independent variable (X) is significantly related to the
mediator variable (M). This is denoted as the a path.
3 Demonstrate that the mediator variable (M) is significantly related to the
dependent variable (Y). This is denoted as the b path.
4 Demonstrate that the relationship between the independent variable (X) is
no longer significantly related to the dependent variable (Y) once the pres-
ence of mediator (M) is accounted for. This is denoted as the c’ path.

Following assessment of these four conditions, if the c’ path is substantially reduced

(to a non-significant c’ path coefficient) following the inclusion of the mediating
variable, the model provides evidence of full mediation. If the c’ became substan-
tially reduced after the mediating variable was added, then the model can be said
to provide evidence of partial mediation.
Although Baron and Kenny (1986) provided one of the most cited
approaches, several studies have criticized their approach (e.g., LeBreton et al.,
2009; MacKinnon et al., 2002). For instance, LeBreton et al. (2009) argued
that in order to specify mediation clearly, the model must have no specification
errors and must be set up in a confirmatory, rather than exploratory, manner.
Specification errors include incorrect causal order, incorrect specification of
variable relationships, or failure to include a specific variable within the rela-
tionship. Bridging forward from the overall issue of specification errors, James
and Brett (1984) argued that three conditions must be met for a model to sat-
isfy the requirements of a confirmatory model: (a) no specification errors, (b)
meaningful relationships between hypothesized variables, and (c) empirically
demonstrated good fit of a model.
 Application of Structural Equation Model 37

The model proposed by Baron and Kenny (1986) does not meet the require-
ments set forth by James and Brett (1984) for two reasons: first, the overall model
and allowances for “partial mediation” do not test or uphold a confirmatory
model; second, the availability of a “partial mediation” designation automatically
includes the assumption that the model is not complete because all related vari-
ables are not included within the model (LeBreton et al., 2009). That is, because
the model provided in Baron and Kenny (1986) does not require hypotheses
regarding full vs. partial mediation, and a finding of a partial mediation insinuates
that there are, in fact, missing variables, it is not a confirmatory model. In addi-
tion, MacKinnon et al. (2002) tested the model of the causal steps against two
other models (including the one from Baron and Kenny) for mediation effects.
They found no optimal approach, but they did find that the Baron and Kenny
model fared the worst in all simulations, primarily due to reduced power.
Lastly, LeBreton et al. (2009) argued that because of an overall lack of causal
modeling within the causal steps procedure, particularly the lack of falsifiable
hypotheses and missing variables, the Baron and Kenny (1986) procedure is not
thorough enough to draw steadfast conclusions regarding causal relationships
and analyses within an SEM framework. Suggesting that a heuristic framework
would better suit such conclusions, LeBreton et al. (2009) recommended an SEM
framework for analyzing mediator relationships, particularly as they become more
complex, for it can accommodate more complex regressions (e.g., multiple IVs,
DVs, mediators) and non-linear relationships without substantial reduction in
power. Another important reason is that mediation is by nature an investigation
of causes and, thus, should be tested using confirmatory techniques such as SEM,
which also directly addresses measurement errors and uses manifest and latent
variables. SEM also allows for testing an entire set of equations in one model and
provides individual parameters and goodness of fit levels. LeBreton et al. (2009)
provided a heuristic to help assess such effects based on two dimensions: (a) how
complex a model is and (b) how speculative (vs. conventional) the empirical
background for such a model is. These two dimensions yield four types of models:
(a) simple speculative, (b) simple conventional, (c) complex speculative, and (d)
complex conventional. LeBreton et al. (2009) suggested that researchers should
analyze either observed variables or latent variables when using simple models,
but not both. Concerning the level of empirical support for a model, they argued
that only OLS estimation procedures be utilized when examining speculative
models, as specification errors are more impactful when using FIML procedures.
On the other hand, researchers can analyze both observed and latent variables
when working with complex conventional models and use FIML-based proce-
dures. Finally, LeBreton et al. (2009) noted the importance of letting theory guide
mediation analyses, including a priori prediction of partial versus full mediation.
In addition to the methods displayed in the summary table, other methods
such as SEM-based mediation analysis and bootstrapping (e.g., Bollen & Stine,
1990; Byon et al., 2020; MacKinnon et al., 2004) have recently become popular.
Bootstrapping is a non-parametric method based on resampling with replacement
38 K. K. Byon et al.

Table 2.4 M
 ediation of the Effects of Attitude, Subjective Norm, and Team
Identification on Visit Intention through the Desire to Visit (n = 200)

Bootstrap Estimate 95% Confidence Interval

Path Coefficients SE Lower Upper

ATT → Des → Int 0.313* 0.069 0.310 0.920

SN → Des → Int 0.194* 0.060 0.115 0.386
TID → Des → Int 0.075 0.066 −0.081 0.348

Note: Attitude toward attending sport event; SN = Subjective norm for attending sport
event; TID = Team identification; Des = Desire to visit; Int = Intention to visit.
* p < .001.

and is performed many times (e.g., 5,000 cycles). The bootstrapping method is an
alternative to Sobel’s method, in which high Type I error rates and low statistical
power with small sample sizes have been identified (MacKinnon et al., 2004).
Bootstrapping controls for Type I errors has statistical power due to its data resa-
mpling procedure (MacKinnon et al., 2004). Table 2.4 shows an example of how
to report the mediation test results in a table.

Moderation
Moderation is a third variable (or construct) that changes the relationship
between two related variables (or constructs) (Hair et al., 2010). Moderation
analysis advances knowledge as testing boundary conditions for relationships
between variables is one of the causal conditions put forth by James et al. (1982).
A moderation effect occurs when the relationship between two variables, X and
Y, varies based on the level of a third variable, Z. Common practice to exam-
ine a moderating effect using a regression follows the following steps. First, X
and Z are centered by subtracting their respective means, called mean-center-
ing. These values are then multiplied to create a cross-product, XZ. Next, Y is
regressed onto X and Z. Then, the product term is added, and the researcher
examines the significance of the product term and the significance of an F
test between the model R-squares. If these values are significant, an interaction
effect is present. Consequently, simple slopes are plotted and tested to examine
the exact nature of the interaction.
While the regression approach is certainly more common, moderation can be
tested using latent interactions in SEM. While these procedures are more tech-
nical, they carry the advantages of the SEM framework. Particular importance is
the ability of SEM to deal with non-normal data. For instance, Edwards (2009)
recommended a bootstrapping procedure to handle interaction effects, which are,
by definition, non-normal. In fact, Cortina et al. (2001) provided an informa-
tive review of the technical and theoretical considerations to be addressed in
modeling latent variable interactions and explained several viable methods for
doing so. They addressed several important issues. First, these types of analyses
 Application of Structural Equation Model 39

must address that multiplicative terms are, by definition, non-normal. Second, in

regression, the interaction term is created simply by multiplying the two mani-
fest variables. This interaction becomes more complex in SEM, where multiple
indicators exist, and multiple potential combinations can be used to create the
product term. Third, how to specify the interaction term – as either exogenous
or endogenous – is sometimes unclear. This question is complicated when, for
instance, the predictor is endogenous, and the moderator is exogenous (as in a
second-stage moderated model). Finally, the addition of interaction terms might
cause identification issues in SEM.
Cortina et al. (2001) reviewed several methods for testing latent variable inter-
actions, all of which are more complex than moderated regression procedures but
vary in technicality and elegance. Researchers might choose different methods
based on the types of data or types of questions they are interested in. Cortina
et al. (2001) conceded that ML is not a panacea but is effective (and robust) under
many conditions. However, in the presence of multivariate non-normality, the
Satorra–Bentler correction may also be used. Even this method represents an
improvement over the regression-based approach, which is unable to use a robust
estimation for the interaction term, which is non-normal in that case as well. In
sport management, a similar procedure was conducted by Trail et al. (2012), where
they used the structural model approach from Jöreskog (2000) to test the latent
interaction effect in an SEM. The latent interaction scores were computed after
saving the latent variable scores through LISREL 8.80, and they were included in
the path model to predict latent dependent variables (Trail et al., 2012). Also, Yim
and Byon (2018) examined the moderating effect of identification with the team
between the two types of satisfaction and consumption intention using latent
moderated SEM (LMS) suggested by Klein and Moosbrugger (2000).
SEM analysis has the ability to capture group comparisons, whether between
different groups of individuals (e.g., different cultural backgrounds) or the same
group of individuals at different times (e.g., team identification levels throughout
a season). However, before concluding group differences or modeling changes,
researchers should examine the assumption of measurement invariance across
groups or occasions (i.e., time points; Zhang, Byon, Xu, et al., 2020). While
this step is often ignored, establishing invariance strengthens study design and
lends credibility to the conclusions derived from the results. Only then can the
researcher know whether meaningful comparisons have been made. Vandenberg
and Lance (2000) developed a straightforward eight-step procedure for examining
invariance in just about every aspect of a measure. Since then, however, theoreti-
cal advances have shown that all eight steps might not be necessary; instead, two
steps are important in almost every context. The decision to use the rest depends
on the research questions. The following sections will discuss each step of the
original framework of Vandenberg and Lance (2000), and recently, amendments
were made by Vandenberg (personal communication, 2013), and how Ployhart
and Oswald (2004) built upon this framework to ask different types of research
questions.
40 K. K. Byon et al.

Omnibus Test
This first test generally measures the invariance of the covariance matrices
(Σ) in total. At the time of Vandenberg and Lance (2000), scholars believed
that if this test revealed no lack of invariance, subsequent tests were unneces-
sary. However, this test has proved problematic, likely due to high Type I errors.
Whatever the cause, subsequent tests sometimes reveal a lack of invariance even
when the omnibus test is non-significant. A significant omnibus result might also
yield a lack of invariance in later steps. Thus, this test is no longer necessary or
sufficient for establishing invariance.

Configural Invariance
Configural invariance is a test of the same fixed and free factor across groups.
This type of invariance is arguably the most important because it establishes that
the measure functions the same way psychologically across different groups. If
comparisons are made when there is a lack of configural invariance, a researcher
is truly comparing apples and oranges. Given a lack of configural invariance, the
researcher should inspect the CFAs to see whether a problematic item might be
removed without severely altering the measure. However, more significant prob-
lems indicate that the researcher should go no further.

Metric Invariance
This test examines whether the values in the ΛX matrix (i.e., the factor load-
ings) are equal across groups. Invariance in this respect indicates that respondents
from different groups are interpreting items in the measure on the same scale
so that their responses might be comparable. Some debate about partial metric
invariance remains; that is, some researchers have suggested that if one item is
invariant, it can be extrapolated to other items. Vandenberg (personal commu-
nication, 2013) suggested that partial invariance might not exist and might be
a slippery slope. Instead of transforming the items, omitting a problematic item
might be recommended, but omitting any more than one might alter the scale.
If the researcher’s goal is to run a path model, for instance, configural and metric
invariance are sufficient.

Scalar Invariance
Scalar invariance is a test of the item intercepts across groups. When examin-
ing group differences (described in more detail below), the researcher should
first constrain the τs of like items to be equal across groups. Doing so will
effectively push the differences up to the latent means, which can then be
tested.
 Application of Structural Equation Model 41

Invariance of Item Uniqueness

Vandenberg and Lance (2000) used this test of the equivalence of the unique-
nesses (Θδ and/or Θε) to indicate the reliability of an item. However, invariance
in this respect could represent true differences.

Equivalence of Factor Variances

Vandenberg and Lance (2000) treated this comparison of factor variances in
the diagonal of the psi matrix as a necessary condition for interpreting item
uniquenesses as reliable and as a secondary test of metric invariance. On the
other hand, Ployhart and Oswald (2004) used this test to measure the homoge-
neity of variance, which is necessary for conducting group comparisons.

Equivalence of Factor Covariances

This test is often conducted simultaneously with the abovementioned step to
establish equal covariances between the latent factors across groups. However,
this test is somewhat redundant with configural invariance, which is a more strin-
gent test and is not often used.

Latent Factor Means

This test examines the differences in latent factor means by forcing equivalence
in the kappa matrix. Again, this step is not often used. Researchers interested
in latent mean differences might prefer the procedure outlined by Ployhart and
Oswald (2004).
Ployhart and Oswald (2004) built on this framework to ask research questions
using mean and covariance structure (MACS) analysis. They addressed three
questions: (a) Is the measure functioning in the same way across groups? (b)
Is there homogeneity of variance? And (c) Are there differences in the latent
means? First, a researcher would conduct tests of configural and metric invari-
ance, as outlined above, in order to ensure that the comparisons will be mean-
ingful. Having established equality in those respects, if mean differences are
hypothesized, the researcher would expect to find a poor fit for scalar invariance
when the item intercepts are constrained to be equal, pushing the differences up
to the latent means. Factor variance equivalence must also be established
using the procedure above. As Ployhart and Oswald (2004) recommended, the
researcher should then fix one of the groups’ means equal to zero; the theory
might guide how this referent group is defined. The mean differences can be
directly tested such that a significant mean would indicate a difference between
that group and the referent group. This approach builds on the invariance steps
by Vandenberg and Lance (2000) to introduce a strategy for hypothesis testing
42 K. K. Byon et al.

in SEM. A similar test could be done with ANOVA or t-tests, but this approach
takes advantage of the ability to model error in SEM and represents a more
informative test.
After the series of CFA invariance tests, Byrne’s (2012) method for SEM invar-
iance is suggested. For the SEM invariance test, the configural model and the
constrained equal model (i.e., invariant factor loading, intercept, and structural
regression paths) are compared, and when the configural model demonstrates a
significantly smaller chi-square value than the constrained model, the moderat-
ing effect is present. Testing for moderating effect using invariance analysis is not
as widely adopted as it should be, considering the increasing interest of research-
ers in sport management to understand group differences (e.g., loyal fans vs. non-
loyal fans, millennials vs. baby boomers, retail customers vs online customers) for
effective marketing of sport products and services. A series of invariance tests are
demonstrated in the illustrative examples at the end of this chapter. The LMS
approach (Klein & Moosbrugger, 2000) is a more advanced method that can
examine the moderation effect of a latent variable. In contrast, the multi-group
SEM analysis after invariance tests can only examine the categorical variables.
LMS uses the full information ML approach based on expectation maximiza-
tion (EM), an approach also known as maximum likelihood robust error (MLR).
Hence, it does not require creating a product term out of the observed variables.
This method also deals with non-normality through a sandwich estimator to
compute the standard errors, similar to bootstrapping.

An Illustrative Example
This illustrative example provides readers with an example of SEM application pro-
cedures and reporting of SEM results. The purpose of this example was to investi-
gate the structural relationships of millennial sport fans’ attitude toward sport game
attendance behavior, the subjective norm for sport game attendance behavior, team
identification, desire to visit sport event, and intention to visit sport event while
taking into consideration the moderating effect of generational differences between
Millennials and Generation X. In the current illustrative example, we developed a
model based on Yim and Byon’s (2020) SFMGB. For the demonstration purpose,
we only selected the following variables: (a) attitude toward sport game attendance
behavior, (b) subjective norm for sport game attendance behavior, (c) team identifi-
cation, (d) desire to visit the game, and (e) intention to visit the game from SFMGB
and included into this illustrative example.
For the purpose of this example, the following six hypotheses were tested: (H1)
attitude will positively influence visit desire, (H2) subjective norm will positively
influence visit desire, (H3) team identification will positively influence visit desire,
(H4) team identification will positively influence visit intention, (H5) desire will
positively influence visit intention, and (H6) generational differences will be
found between millennial sport fans and Generation X sport fans when predict-
ing intention to attend sport event (Figure 2.3). To demonstrate, the proposed
 Application of Structural Equation Model 43

Figure 2.3 Proposed Research Model and Hypotheses.

model is the most plausible model among theoretically equivalent models, it was
compared with two rival models (see Figure 2.2). The model B (alternative model 1
where the team identification was theorized to predict the desire to visit the game
only) was based on Perugini and Bagozzi’s (2001) MGB, where they theorized and
empirically found that the desire as a mediator between motivational variable and
behavioral intention. Previous researchers have also found that team identifica-
tion predicts sport fans’ behavioral intention (e.g., Fink et al., 2009); therefore,
model C (alternative model 2 where the team identification only predicts the
intention) was compared.
Using Amazon Mechanical Turk, a total of 400 data (i.e., 200 Millennials and
200 Generation X) were collected, which were subjected to an SEM. The SEM
model includes three exogenous latent variables (i.e., attitude toward sport game
attendance behavior, subjective norm for sport game attendance behavior, team
identification) and two endogenous latent variables (i.e., desire and attendance
intentions). Information on generation (i.e., Millennials and Generation X) was
collected to examine the moderation effect.
Anderson and Gerbing’s (1988) two-step approach was used to test the meas-
urement model via CFA and then test the hypothesized structural model using
SEM procedures. For this illustrative example, we used a Mplus 7.0. We followed
Hair et al.’s (2010) suggestion to use multiple goodness of fit measures, includ-
ing chi-square statistic (χ2), normed chi-square (χ2/df), RMSEA, SRMR, TLI, and
CFI. For the chi-square statistic (χ2), it is expected to have a non-significant dif-
ference. Bollen (1989) suggested that cut-off values of less than 3.0 for the normed
44 K. K. Byon et al.

chi-square are considered a reasonable fit. According to Hu and Bentler (1999),

the RMSEA value of .06 would indicate a close fit, and RMSEA between .06
and .08 indicates an acceptable fit. While smaller values of SRMR show good fit,
values less than .10 are generally considered acceptable fit (Kline, 2011). TLI and
CFI values greater than .90 indicate an acceptable fit, and values larger than .95
show a close fit (Kline, 2011).
To determine convergent validity, the researcher evaluated the magnitude and
statistical significance of factor loadings for each observed variable. Generally, fac-
tor loadings should be statistically significant (Anderson & Gerbing, 1988), and the
magnitude of factor loadings should be at least .50 or higher, ideally, .70 or higher
(Hair et al., 2010). To establish discriminant validity, two methods were employed:
(a) examination of the interfactor correlations (Kline, 2011), where discriminant
validity can be established when interfactor correlation is below .85; and (b) Fornell
and Larcker’s (1981) method, which stipulates that a squared correlation between
two constructs should be lower than the AVE for each construct. To measure the
scale’s reliability, three tests, including Cronbach’s alpha values, construct reliability
(CR), and average variance extracted (AVE), were employed. A cut-off value of .70
was adopted to determine the internal consistency of the Cronbach’s alpha and
CR (Fornell & Larcker, 1981; Hair et al., 2010). We adopted AVE .50 or greater
suggested by Hair et al. (2010). Construct validity was assessed by means of test-
ing convergent validity and discriminant validity. Once the measurement model
was confirmed to be of good psychometric properties, we proceeded with an SEM
analysis to examine the hypothesized structural relationships among the factors.
The same fit index criteria were employed to examine the structural model as with
the measurement model. Path coefficients were used to determine the direct and
indirect relationships among the various sets of factors.
Prior to estimating the measurement model and structural model, various
assumption tests were conducted that included normality. In terms of the data
normality, skewness and kurtosis for the items were examined to evaluate the
normality of variables. An absolute value of 3.0 was adopted as the cut-off value
for skewness and kurtosis (Chou & Bentler, 1995). The findings revealed that
all skewness and kurtosis values were well within the acceptable threshold. For
instance, the skewness values ranged from −2.487 to −0.347. The kurtosis values
ranged from −1.331 to 2.020.
The overall model fit of the measurement model was acceptable (χ2 = 170.288,
df = 67, p < .001; χ2/df = 2.54; RMSEA = .088, 90% CI = .072−.104; SRMR = .047;
CFI = .96; and TLI = .95). Convergent validity was satisfactory, where all indicator
loadings were statistically significant and well above the minimum criterion of .50
(Anderson & Gerbing, 1988; Hair et al., 2010). None of the interfactor correla-
tions were above the suggested threshold of .85 (Kline, 2011), ranging from .38
(between team identification and intention) to .71 (between attitude and desire),
indicating good discriminant validity (see Table 2.1). As a result of applying the
Fornell and Larcker’s (1981) method, it was found that none of the squared cor-
relations between any of the two constructs were above the AVE value of the
respective construct, indicating strong discriminant validity of the measurement
 Application of Structural Equation Model 45

model. Next, Cronbach’s alpha, CR, and AVE values of all factors were examined.
The measurement model was of excellent reliability as all the values exceeded the
above-mentioned cut-off criteria. Obtaining satisfactory psychometric properties
of the measurement model, it was deemed appropriate to proceed with testing the
hypothesized structural model.
Before estimating path coefficients for the hypothesized structural model, good-
ness of fit indexes for the structural model was first evaluated. The overall model
fit was reasonably well (χ2 = 181.71, df = 69, p < .001; χ2/df = 2.63; RMSEA = .090,
90% CI = .074– .106; SRMR = .057; CFI = .96; and TLI = .95). The chi-square
difference test (between measurement and structural regression model following
Anderson and Gerbing’s 1988 two-step approach) was found to be statistically
significant (Δχ2(2) = 11.422; p < .01) but when comparing the loadings estimates
between CFA and SEM to check for interpretational confounding (Mesquita &
Lazzarini, 2008), they were deemed to be similar.
Two competing (alternative) models were developed to examine the adequacy
of the proposed model to the data. While in the proposed model, team iden-
tification predicted both desire and intention, in the competing models, team
identification predicted either only desire (model B) or intention (model C)
(see Figure 2.2). To compare the models, in addition to the model fit indices com-
parison, we conducted a chi-square difference test since the competing models
were nested models. The proposed model (model A) showed a significantly better
model fit than model B (Δχ2(1) = 5.86; p < .05) in the chi-square difference test,
so we chose the proposed model over the model B. The chi-square difference test
between the proposed model and model C was non-significant (Δχ2(1) = 3.21; p =
non-significant); however, when examining the model fit indices and R2 values,
the proposed model was accepted as the final plausible model (see Table 2.3).
As shown in Table 2.2, Hypothesis 1, which hypothesized an influence of atti-
tude on the desire to attend, was supported (β = .512, p < .001). Hypothesis 2 for
predicting an influence of subjective norm on desire was supported (β = .318, p <
.001). However, Hypothesis 2, which posited an influence of team identification
on desire, was not supported. Hypothesis 4 for predicting team identification’s
influence on intentions was supported (β = .177, p < .05). Hypothesis 5, which pos-
tulated an influence of desire on intention was supported (β = .518, p < .001). One
of the aims of this illustrative example was to examine the moderating effect of
generation (i.e., Millennial vs. Generation X) in predicting game visit intentions. A
multi-group SEM was conducted to detect the moderating impact of generation.
In order to conduct group comparisons, a series of multi-group CFA invariance
tests were conducted by following the recommended procedures from Vandenberg
and Lance (2000) and Ployhart and Oswald (2004). For the CFA invariance test
between Millennials and Generation X for game visit intention, the factor vari-
ance and covariance model (χ2 = 411.27; df = 171; CFI = .95; TLI = .95; RMSEA =
.084; and SRMR = .084) and invariant factor means model (χ2 = 423.56; df = 176;
CFI = .95; TLI = .95; RMSEA = .084; and SRMR = .097) were compared. The
chi-square difference test was found to be statistically significant (Δχ2(5) = 12.29;
p < .05), indicating that there was a moderation effect (see Table 2.5).
46 K. K. Byon et al.

Table 2.5 T
 est for CFA Invariance of Millennials and Generation X for Game Attending
Behavior: Summary of Model Fit and χ -Difference-Test Statistics (N = 400)
2

Model
Model ML χ 2 df SRMSR RMSEA TLI CFI Comparison Δ ML χ 2 Δdf p
Configural model 310.32 134 0.042 0.081 0.95 0.97 – – – –
Metric 319.62 143 0.050 0.079 0.96 0.97 vs. Configural 9.30 9 ns
invariance
Uniqueness 398.41** 157 0.057 0.088 0.94 0.95 vs. Metric 78.79 14 <.001
invariance
Scalar test 416.79 171 0.066 0.085 0.95 0.95 NA
Means different 404.43 166 0.060 0.085 0.95 0.95 vs. Uniqueness 6.02 9 ns
Factor variance 411.27 171 0.084 0.084 0.95 0.95 vs. Mean 6.84 5 ns
and covariance different
Invariant factor 423.56* 176 0.097 0.084 0.95 0.95 vs. Factor var. 12.29 5 <.05
means

Note: ns = non-significant.
** p < .001.
* p < .05.

After revealing the moderation effect of generation, a multi-group SEM was

conducted to test Hypothesis 6. The goodness of fit indices of the structural
model were examined, and it was found to fit the data well (χ2 = 339.45; df = 156;
CFI = .96; TLI = .96; RMSEA = .077; and SRMR = .058). The results of the
multi-group SEM showed differences between the Millennials and Gen X (see
Figure 2.4). Both Millennials’ and Gen X’s attitudes significantly predicted
the desire to visit (Millennial: β = .52, p < .001; Gen X: β = .56, p < .001); however,

Figure 2.4 Multi-group SEM Results.

Note. *p < .05. **p < .001.
 Application of Structural Equation Model 47

subjective norm was only significant among the Millennials (Millennial: β = .31,
p < .001; Gen X: β = .03, p = non-significant) while only Gen X’s team identi-
fication significantly predicted the desire to visit (Millennial: β = .12, p = non-
significant; Gen X: β = .38, p < .001). This result implies that for the Gen X fans,
team identification was an important antecedent to form a desire, but for the mil-
lennial sport fans, the subjective norm, such as peer influence, was a significant
predictor for the desire to visit. Both groups’ team identification and the desire to
visit predicted the intention to visit.

References
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103(3), 411–423.
Arbuckle, J. L. (2006). AMOS 7.0 user’s guide. Chicago, IL: SmallWalters Corporation.
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.
Bollen, K. A. (1989). Structural equations with latent variables. New York, NY: Wiley, 1989,
c1988.
Bollen, K. A., & Stine, R. (1990). Direct and indirect effects: Classical and bootstrap
estimates of variability. Sociological Methodology, 20, 115–140.
Breckler, S. J. (1990). Applications of covariance structure modeling in psychology: Cause
for concern? Psychological Bulletin, 107, 260–273.
Brown, T. A. (2015). Confirmatory factor analysis for applied research (2nd edition.). New
York, NY: The Guilford Press.
Byon, K. K., Salge, C., Baker, T. A., & Jones, C. W. (2020). Team identification and
negotiation: A mediated-moderation model of constraints, motivation, and sport con-
sumption. Sport Marketing Quarterly, 29(4), 282–295.
Byon, K., Zhang, J. J., & Baker, T. (2013). Impact of core and peripheral service quality on
consumption behavior of professional team sport spectators as mediated by perceived
value. European Sport Management Quarterly, 13(2), 232–263. doi:10.1080/16184742.
2013.767278
Byon, K. K., Zhang, J. J., & Connaughton, D. P. (2010). Dimensions of general mar-
ket demand associated with professional team sports: Development of a scale. Sport
Management Review, 13(2), 142–157.
Byon, K. K., Zhang, Y, Hsu, N. Y., Drane, D. D., Pitts, B. G., & Zhang, J. J. (2018).
General game support programs associated with professional team sports. In J. J. Zhang
& B. G. Pitts (Eds.), The global football industry: Marketing perspectives (pp. 303–330).
London, UK: Routledge.
Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts, applications,
and programming. New York, NY: Routledge/Taylor & Francis Group.
Chou, C. P., & Bentler, P. M. (1995). Estimates and tests in structural equation modeling.
In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues and applications
(pp. 37–55). Thousand Oaks, CA: Sage.
Cortina, J. M., Chen, G., & Dunlap, W. P. (2001). Testing interaction effects in LISREL:
Examination and illustration of available procedures. Organizational Research Methods,
4, 324–360.
48 K. K. Byon et al.

Edwards, J. R. (2009). Seven deadly myths of testing moderation in organizational

research. In C.E. Lance & R.J. Vandenberg (Eds.), Statistical and methodological myths
and urban legends: Received doctrine, verity, and fable in the organizational and social sciences
(pp. 143–164). New York, NY: Routledge.
Fink, J. S., Parker, H. M., Brett, M., & Higgins, J. (2009). Off-field behavior of athletes
and team identification: Using social identity theory and balance theory to explain fan
reactions. Journal of Sport Management, 23, 142–155.
Finney, S. J., & Distefano, C. (2006). Non-normal and categorical data in structural equa-
tion modeling. In G. R. Hancock & R. O. Mueller (Eds.), Structural equation modeling:
A second course (pp. 269–314). Greenwich, CT: IAP.
Fornell, C., & Larcker, D. F. (1981). Structural equation models with unobservable var-
iables and measurement error: Algebra and statistics, Journal of Marketing Research, 1,
382–388.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis.
Upper Saddle River, NJ: Prentice Hall International.
Hoyle, R. H., & Panter, A. T. (1995). Writing about structural equation models. In R. H.
Hoyle (Ed.), Structural equation modeling: Concepts, issues, and applications, (pp. 158–176).
Thousand Oaks, CA: Sage.
Hu, L., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity
to under parameterized model misspecification. Psychological Methods, 3(4), 424–453.
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), 1–55.
James, L. R., & Brett, J. M. (1984). Mediators, moderators, and tests for mediation. Journal
of Applied Psychology, 69(2), 307–321.
James, L. R., Mulaik, S. A., & Brett, J. M. (1982). Causal analysis: assumptions, models, and
data. Beverly Hills, CA: Sage Publications.
Jöreskog, K. G. (2000). Latent variable scores and their used. Lincolnwood, IL: Scientific
Software International, Inc.
Jöreskog, K.G., & Sörbom, D. (1996). PRELIS 2: User’s reference guide [Computer software
manual]. Chicago, IL: Scientific Software International.
Klein, A., & Moosbrugger, H. (2000). Maximum likelihood estimation of latent interac-
tion effects with the LMS method. Psychometrika, 65(4), 457–474.
Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New
York, NY: Guilford.
Kwak, D. H., McDaniel, S., & Kim, K. T. (2012). Revisiting the satisfaction-loyalty rela-
tionship in the sport video gaming context: The mediating role of consumer expertise.
Journal of Sport Management, 26(1), 81–91.
LeBreton, J. M., Wu, J., & Bing, M. N. (2009). The truth(s) on testing for mediation.
In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and
urban legends: Received doctrine, verity, and fable in the organizational and social sciences
(pp. 107–141). New York, NY: Routledge.
Lee, Y., Yim, B. H., Jones, C. W., & Kim, B. (2016). The extended marketing mix in the
context of dance as a performing art. Social Behavior and Personality: An International
Journal, 44(6), 1043–1059.
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002).
A comparison of methods to test mediation and other intervening variable effects.
Psychological Methods, 7(1), 83–104.
 Application of Structural Equation Model 49

MacKinnon, D. P., Lockwood, C. M., & Williams, J. (2004). Confidence limits for the
indirect effect: Distribution of the product and resampling methods. Multivariate
Behavioral Research, 39(1), 99–128.
Magnusen, M., Kim, J. W., & Kim, Y. K. (2012). A relationship marketing catalyst: the
salience of reciprocity to sport organization–sport consumer relationships. European
Sport Management Quarterly, 12(5), 501–524.
Maxwell, S. E., & Cole, D. A. (1995). Tips for writing (and reading) methodological arti-
cles. Psychological Bulletin, 118, 193–198.
Mesquita, L. F., & Lazzarini, S. G. (2008). Horizontal and vertical relationships in
developing economies: Implications for SMEs’ access to global markets. Academy of
Management Journal, 51(2), 359–380.
Meyers, L. S., Gamst, G., & Guarino, A. J. (2016). Applied multivariate research: Design and
interpretation (2nd ed.). Thousand Oaks, CA: Sage
Perugini, M., & Bagozzi, R. P. (2001). The role of desires and anticipated emotions in
goal-directed behaviours: Broadening and deepening the theory of planned behaviour.
British Journal of Social Psychology, 40(1), 79–98.
Ployhart, R. E., & Oswald, F. L. (2004). Applications of mean and covariance structure
analysis: Integrating correlational and experimental approaches. Organizational Research
Methods, 7(1), 27–65.
Raykov, T., Tomer, A., & Nesselroade, J. R. (1991). Reporting structural equation mod-
eling results in Psychology and Aging: Some proposed guidelines. Psychology and Aging,
6, 499–533.
Ryu, E. (2011). Effects of skewness and kurtosis on normal-theory based maximum
likelihood test statistic in multilevel structural equation modeling. Behavior Research
Methods, 43(4), 1066–1074.
Satorra, A., & Bentler, P.M. (2001). A scaled difference chi-square statistics for moment
structure analysis. Psychometrika, 66, 507–514.
Schumacker, R. E., & Lomax, R. G. (2010). A beginner’s guide to structural equation mode-
ling. New York, NY: Routledge.
Tanaka, J. S. (1993). Multifaceted conceptions of fit in structural equation models. Sage
Focus Editions, 154, 10–10.
Trail, G., Kim, Y., Kwon, H., Harrolle, M., Braunstein-Minkove, J., & Dick, R. (2012).
The effects of vicarious achievement on BIRGing and CORFing: Testing moderating
and mediating effects of team identification. Sport Management Review, 15(3), 345–354.
Vandenberg, R. J., & Grelle, D. M. (2009). Alternative model specifications in struc-
tural equation modeling: Facts, fictions, and truth. In C. E. Lance, & R. J. Vandenberg
(Eds.), Statistical and methodological myths and urban legends: Doctrine, verity and fable in
the organizational and social sciences (pp. 165–191). New York, NY: Routledge.
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(1), 4–69.
Walker, M., & Heere, B. (2011). Consumer attitudes toward responsible entities in sport
(CARES): Scale development and model testing. Sport Management Review, 14(2),
153–166.
Wright, B. K., & Williams, A. S., & Byon, K. K. (2017). Brand marketing via Facebook:
An investigation of the marketing mix, brand equity, and purchase intention in the
fitness industry. Marketing Management Journal, 27(2), 131–142.
50 K. K. Byon et al.

Yim, B. H., & Byon, K. K. (2020). Critical factors in the sport consumption decision
making process of millennial sport fans: A revised model of goal-directed behavior.
International Journal of Sports Marketing and Sponsorship, 21(3), 427–447.
Yim, B, H., Byon, K. K., & Baker, T. A. (2018). The mediating role of game and service
satisfaction between fan emotion and behavioral intentions: Examination in winning
and losing situations. Sport Marketing Quarterly, 27(2), 93–106.
Yoshida, M., & Gordon, B. (2012). Who is more influenced by customer equity drivers?
A moderator analysis in a professional soccer context. Sport Management Review, 15(4),
389–403.
Zhang, J. C., Byon, K. K., Tsuji, Y., & Pedersen, P. M. (2020). Co-created value influ-
ences residents’ support toward the sporting event through the mediating mechanism
of gratitude. European Sport Management Quarterly. Advanced online publication.
doi: 10.1080/16184742.2020.1836011
Zhang, J. C., Byon, K. K., Xu, K., & Huang, H. R. (2020). Event impact on residents’ satis-
faction and behavioral intention of the mega sporting event: Pre-post study. International
Journal of Sports Marketing and Sponsorship, 21(3), 487–511.
Zhang, J. J., Lam, E, T, C., Smith, D. W., Fleming, D. S., & Connaughton, D. P. (2006).
Development of the scale for program facilitators to assess the effectiveness of after
school achievement programs. Measurement in Physical Education and Exercise Science,
10(3), 151–167.
Zhang, Y., & Byon, K. K. (2017). Push and pull factors associated with the CTTSL game
events between on-site and online consumers. International Journal of Sports Marketing
and Sponsorship, 18(1), 48–69.
Chapter 3

Disruptive Technologies on
Sport Event Marketing
New Centrality of Digital Communication
Patrizia Zagnoli and Elena Radicchi

Introduction
The new complexity of the sport industry requires a growing level of investments
to be competitive not only in athletic performance but also in the sport busi-
ness. This leads sports organizations to establish a wide composite “network of
collaborative relationships” (Normann & Ramirez, 1993; Lusch & Vargo, 2006)
with different stakeholders (Freeman, 1984; Freeman et al., 2007) for the imple-
mentation of various sport events such as matches, tournaments, games, etc., that
can be at the national, international, and global level. The variety of stakehold-
ers converges to let the sport event take place at a different level of complexity
(amateur, professional, etc.). Sport events imply systemic relations among multiple
actors involved in specific transactions able to collect and exchange economic,
financial, technological resources toward an organizational frame where skills
and capabilities are managed.
Sport clubs need to gain specific resources like talented players, venues, facil-
ities, sporting goods, and equipment. Sport organizations must raise funds, gain
revenues, and cover costs to possibly reach both a positive sportive and financial
performance. This entails offering an exciting on-field spectacle, valuable services
within stadiums and arena, additional activities (merchandising, restoration,
social media services, connection, appliances, etc.) to enhance fans’ identifica-
tion with an athlete or a team, strengthen brand loyalty, and increase the media
audience. A critical dimension for sport clubs is the capability to exchange dif-
ferent kinds of services with multiple actors like sponsors, TV networks, sporting
equipment suppliers, institutions, marketing agencies, etc.
For global corporations, sport is very attractive to promote and sell their prod-
ucts. In recent years, they have increasingly secured ownership and control over
internationally well-known soccer clubs, Formula 1, motorbikes, and cycling
teams. On the one hand, buying into a major sport franchise is a business invest-
ment that might increase the owners’ “trophy status” (Foster et al., 2005) and
prestige. It can also generate financial gains and launch investors’ products or
brands into new markets. On the other hand, corporations play a strategic role in
the development of sport organizations: not only do they integrate their financial

DOI: 10.4324/9781003302476-3
52 P. Zagnoli and E. Radicchi

resources providing funds to the team, but they contribute with their strategic
and managerial skills in operating them. Company support allows sport organi-
zations to achieve financial, human, and managerial resources, capabilities, and
“complementary assets” (Pisano, 2006; Teece, 1986; Teece et al., 1997).
Sporting equipment suppliers capitalize on sport events as well. Sporting goods
manufacturers and distributors supply professional clubs with technologies, mate-
rials, and components necessary to the sport performance outcome. Typically,
soccer balls, shoes, fitness equipment, training software, etc. Sport is a great arena
for R&D processes and product testing. Sporting technical partners are increas-
ingly becoming sponsors of teams and events: companies gain many opportuni-
ties and synergies, such as strengthening of their ties with the local context. They
can also boost their customers’ loyalty and develop a clear brand identity in order
to achieve economic benefits from the investments they have made. Fans and
supporters play a catalytic role in sport. Their passion, excitement, and involve-
ment are crucial for the event’s implementation. Subscribers and spectators are
key supporters in producing the game and making the sport event happen. They
play a meaningful demand function, involving themselves and dragging friends,
colleagues, and families. Fans are direct consumers of the service provided by the
sport club (match). They trigger a virtuous circle where media, suppliers, distribu-
tors, institutions, sponsors, club investors, athletes play complementary roles and
create significant value at the “clue” moment when the sport event takes place
(Zagnoli & Radicchi, 2016). In this complex relational context, seemingly new
social media can appear just like a new way to distribute the event content. Today,
they play a meaningful role in stimulating the development of interactions among
a lot of actors involved within the sport industry.
The new communication channels are a kind of “soundboard” of the sport
content. New social media are strategic promotional tools for sport clubs and
sponsors’ image and brands growth. Moreover, social media allow sharing of sport
content and “voicing” opinions: an important part of the fans’ experience (Pine
& Gilmore, 2011), “in and around the sport event”. The new technological and
fruition context is the base on which eSport is taking off. From arcade games
(pinball machines, electro-mechanical games, redemption games, or merchan-
disers) in the 1980s and local area network (LAN) parties in the early 1990s – a
gathering of people with computers or compatible game consoles, where a LAN
connection is established to play multiplayer video games together – to the rise
of massively multiplayer online games in the 2000s, gamers have spent decades
building the foundations of eSports. The development of robust computing and
graphics platforms, along with the growth of social media and high-speed inter-
net access, has helped enable designers to introduce highly responsive and deeply
immersive shared gaming environments. Now, the market for video games has
grown so much that industry revenues are more than double compared to global
box-office receipts for movies.
The relation between new technologies and sport is therefore not only merely
confined to a distribution role. The new communication channels are really
 Disruptive Technologies 53

a leading engine of the value co-creation process. They stimulate the people’s
potential demand and directly influence the partners’ strategies. Media and new
media companies compete to attract brands from different sectors, thus “multiply-
ing” the spread value of sport contents. The contemporary sport real event is still
the core of the process: new technologies have made the sport event more social,
interactive, and immersive. The development of robust computing and graphics
platforms, along with the growth of social media and high-speed internet access,
has indeed helped enable major leagues and top professional teams to introduce
highly responsive and deeply immersive shared gaming environments across the
real event. This research aims to enlighten the new complexity of organizing and
marketing a sport event in the light of the emergence of new variables such as
the eSport that apparently seems to represent a natural evolution, but, upon close
examination, it emerges like a disruptive technology’s affirmation process.

New Sport Event Marketing Mix

The marketing mix is defined as a set of strategies and activities that involve
Product, Price, Promotion, and Place (distribution). These are commonly referred
to as “The Four Ps”. An additional P (People) is usually considered in the ser-
vices marketing, which is particularly important for the sport sector. The fact that
these elements are grouped into a set or a “mix” is crucial because they should
be coordinated together in an integrated strategic approach. Traditionally, at the
heart of the decision-making process regarding a sport club marketing mix are the
research and knowledge related to the organization of the event. The core prod-
uct is the sport event – defined as the actual game in terms of competition (players
training and sport equipment supply), the sponsor companies, the event venues
and facilities, the ticketing selling process, and the promotion and distribution of
core and additional services. Finally, specific services must be designed to increase
the fans’ attachment and loyalty. All factors must be given careful attention. If
the sport event organizer ignores one or the other, this increases the chances of
making improper decisions (Figure 3.1).
Over the years, the sport event marketing mix has progressively changed. First,
a new important role has been gained by the location where the sport event is
taking place. Place and distribution in marketing terms is the location and the
process of getting the sport product to the consumer. It, therefore, refers to the
location of the sport product (stadium and arena), the point of origin for supply-
ing the “game” to the fans. Factors related to the physical location of the sport
can have multiple favorable or unfavorable effects on the event marketing suc-
cess. To ensure a positive effect, contemporary sport venues are more and more
easily accessible (highway systems, parking, walkways, and ramps) and try to have
an attractive physical appearance in an enjoyable, convenient, and sustainable
environment, with safe and pleasant surroundings. Layout design, facility aes-
thetics, seating comfort, electronic equipment, and displays all significantly influ-
ence how sports fans perceive the stadium experience. In turn, the higher this
54 P. Zagnoli and E. Radicchi

Figure 3.1 The Disruption of the Sport Event Marketing Mix.

perceived quality of the sports venue, the higher the sports fan’s satisfaction and
attachment to their sports club. If sport fans are enjoying the venue’s environ-
ment, they are likely to stay in the stadium for a longer period, and they have a
variety of entertainment and suppliers that attract them to spend more money;
also, they are likely to return.
Distribution is strictly connected and sometimes even overlapped, with
Promotion, a different avenue for raising fans’ awareness. This process may involve
a variety of methods for gaining the attention of users. In sport, promotion cov-
ers a range of interrelated traditional and innovative tools (print media, sponsor-
ship, advertising, PR, e-commerce, TV broadcasting, social network, etc.). All these
instruments are designed to catch and stimulate the interest and the awareness
of fans and spectators, ending to a great desire to participate in the “sport event”.
Promotion is more and more about communicating and engaging people through
a mix of new digital media that have spirally disrupted the sport event marketing
mix, giving digital communication a pivotal role. The new centrality of com-
munication has been enhanced based on new technological digital paradigms.
Figure 3.2 represents the media devices that have supported the evolution of the
technological communication paradigm toward digital disruption.
The process of digitalizing, that is, “to codify, transform, and broadcast infor-
mation as a bit”, in the computer systems at the beginning of the ‘80s has been
the hauling element in the transformation of the telecommunications sector
producing important and unexpected potentials. In this phase, the technologi-
cal paradigm (Dosi, 1988) is characterized by the emergence of a personal com-
puter that allows the database construction and, therefore, the management of
a complex system of information on the PC, also by a single user. In the ‘90s,
 Disruptive Technologies 55

Figure 3.2 Media Devices: Communication Technological Paradigm and Digital Disruption.

the interaction between computer science and telecommunications represents the

embryonic phase of the multimedia paradigm (Zagnoli & Cardini, 1994). The
new technological paradigm is characterized by the Internet, which realizes
the convergence between the computer science sector and telecommunications:
information is transmitted through the web by overcoming space and time limits.
Thanks to the development of the telecommunications sector and its conver-
gence with computer science (ICT), new ways of organizing information emerged,
both in terms of data elaboration and transmission and sharing.
The convergence of computing, telecommunication, and audio-visual technol-
ogy (Yoffie, 1997; Tapscott, 1996) led to the diffusion of new media that is both
interactive and digitally distributed, such as websites and mobile telecommuni-
cation (Santomier & Shuart, 2008). Two fundamental features distinguish new
media from all other traditional media. In the past, each device usually accom-
plished some basic tasks: media contents (concerts, sport events, shows, etc.) were
displayed on a television, music came through a tape or a CD, and videogames
were played through a console. The evolution of ICT technologies changed
the way how content is transmitted and accessed. Multimedia raises together a
great variety of devices: television, radio, PC, laptop, tablet, mobile phones, and
smartphones. Today’s content like matches, music concerts, movies, etc., can be
watched on television, on a tablet through high-speed internet access, or on a
smartphone. From the user’s perspective, all of those have been interchangeable
tools. What makes the difference for the final consumer’s choice is the context,
quality of the image, video, audio, graphics, etc., and the potential of interactivity.
A major aspect of the changes in new media consumption is to provide the
fan with the feeling of a sort of centrality that goes from broadcaster to viewer.
Consumers have the technical potential to become editors themselves, empow-
ered as to which messages they receive (Tapscott et al., 2000). Moreover, new
56 P. Zagnoli and E. Radicchi

media technologies allow increasingly constant connectivity (Mazmanian &

Erickson, 2014). With smartphones set to overtake the desktop as the primary
means by which people browse the Internet, it seems that the need to connect
with each other is something that is no longer confined to a single space and at a
certain time of the day. Compelled to receive constant information updates from
their ever-expanding peer networks, the average person has been transformed
into a “hyper-connected habitué” of social media (Buzzetto & Moore, 2013).
Interactivity – intended as the “interaction” of the user with the new media,
the event, and other users – is becoming central to how people, especially young
people, use new media. In the Digital Economy, people are no longer merely pas-
sive agents but participate interactively (Tapscott, 1996). The emerging new dig-
ital technologies, particularly the broadband Internet and other different ways
of broadcasting (mobile phones, digital television, etc.), enable the spectator to
interact not only with the device but also with other people to create a rela-
tional network that gives place to virtual “communities”. The latest development
of communication and technological paradigm is moving forward due to the
changes brought about by digital disruption such as virtual reality, augmented real-
ity, and eSport.
Virtual reality (VR) is an artificial environment which is experienced through
sensory stimuli (such as sights and sounds) provided by a computer and in which
one’s actions partially determine what happens in the environment. The purpose
of VR is to allow a person to experience and manipulate the environment as if it
were the real world. The best virtual realities can immerse the user completely.
VR should not be confused with simple 3-D environments like those found in
computer games, where you get to experience and manipulate the environment
through an avatar rather than personally becoming part of the virtual world.
Although augmented reality (AR) has gained much research attention in recent
years, the term AR was given different meanings by varying researchers. AR can
be defined as the use of wearable devices that are worn, like smart glasses, through
which it is possible to merge virtual information with physical information in a
user’s view field. In recent years, VR and AR have seen a massive application in
the field of sports, gradually transforming the way fans and athletes enjoy and per-
form different disciplines. The ultimate evolution is eSport, which encompasses a
wide range of digital and virtual activities played by the sport fans.

The eSport Disruption

By 2020, the global eSports market is expected to generate $1.5 billion in annual
revenues, primarily from sponsorships and advertising to an estimated global
audience of 600 million fans (Nielsen, 2017). Marketers presented more than
600 brand sponsorships on eSport titles and events in 2017 alone. Additional
revenues come from betting, ticket sales, and merchandise. Revenues and engage-
ment from eSports are increasingly driving top gaming publishers’ valuation and
sales. For example, in its 2017 earnings announcement, Electronic Arts – the
 Disruptive Technologies 57

videogames company based in Redwood City, California – reported 18 million

players engaged in competitive gaming for its FIFA 18 (soccer) and Madden NFL
(National Football League) 18 titles, up 75 percent over the previous year. More
popular game titles are shifting to offer competitive multiplayer experiences.
For many businesses in media and entertainment, eSports offers a way to reach
a demographic which has been increasingly beyond their grasp. Overall, players
and fans are young, less likely to watch TV, and often less interested in profes-
sional sports than the general population. In 2017, Nielsen Media evaluated
the demographics of eSports enthusiasts and found that 70 percent are male,
aged 13–40, two-thirds said they watch live eSports, and 37 percent have attended
live eSports events. Deloitte’s own digital media trends survey shows that half of
Generation X respondents say they play video games at least once a week, almost
as much as millennials and Generation Z respondents. Generation X leads all gen-
erations in the amount of time they spend playing games on mobile. Older people
might dismiss video games, but their children grew up with them and continued
playing on consoles, smartphones, and PCs (Nielsen, 2017). The development
toward eSports represents a very recent disruptive affirmation. Over the last two
decades, multiplayer competitive video gaming has globalized and spread across
the Internet and social media, aggregating large audiences of digital natives who
have grown up in these virtual environments. As online access minimizes physi-
cal differences between individuals in a growing number of settings, the world of
eSports allows anyone to become a player and even a broadcaster. The potential
of disruptive technology emphasizes the sport content making it the “eSport” in
which the event is in many ways deeply inclusive and virtually immersive.
Disruptive technologies imply that the digital acquisition era is implemented
coupled with AR and eSport fantasy. Disruptive means that we access the stage
of complex digital communication from communication technology to enrich the
real sport event by gamifying the live fan participation. Disruptive technologies
are a great social and economic incentive for stakeholders and especially in pro-
fessional sport are becoming an inner part of the sport event to a point that rising
new buildings are attractive and considered valuable in terms of a high price to
be spent. Fans perceive great participation if they can watch the event and simul-
taneously use new technology to amplify their experience toward “gamification”.
Along the way, digital games have evolved from being narrative-based, single-
player experiences to persistent social platforms with integrated and real-time
communication. This shift to games as a service sees game publishers adding a
variety of contents that users can customize, alongside free applications that keep
the “game fresh”. Some users also stream their play and commentary, bringing
potential marketing opportunities for sports clubs and the sponsors.
The eSports users can represent a huge revenue potential in merchandise and
content. Many teams provide visual appearances, including virtual team jerseys,
that individual gamers can use for their on-screen avatar as downloadable con-
tent through the game platform. As with traditional sports, merchandise can be
physical and digital and utilized to “extend” the franchise and its brand. Gaming
58 P. Zagnoli and E. Radicchi

platforms can also be a space to offer co-branded content, digital accessories,

and additional experiences. Anyway, to exploit the higher worth from these
eSport potentials, teams and companies should be able to offer valuable services
besides deploying rapid-response marketing. For example, in 2017, a large auto-
maker sponsored a tournament at an eSports venue, parking cars out front and
showing off its logo across stage and screen. Fans photographed and “meme-ified”
the pictures, insulting the brand and sharing it across their digital networks.
The carmaker quickly responded by making its own Instagram pictures to mock
itself and the fans. People liked this ironic reaction of the car producer, and the
brand conversation turned positive. Moreover, many fans enjoy watching eSports
together in stadiums, with other players on location and massive screens tracking
the action; championship games easily fill 15,000-seat arenas, and gaming hard-
ware manufacturers are signing deals with venues to be the exclusive providers
of PCs and peripherals. Developers are investing massively in new eSports ven-
ues, training facilities, and living spaces for teams. For example, Super League
Gaming repurposes underused movie theaters as eSports arenas and has raised more
than $28 million from investors, leading some to envision an eSports future for
declining shopping malls.
Regarding its technological development, eSport can appear quite different to
traditional team sports fans. Many eSport games make it challenging for viewers
to focus attention on the play. Indeed, broadcast media may find it difficult to
keep up with many games’ pace and multiple viewpoints; coordinating 30 in-game
camera views into a cohesive broadcast is already requiring heavy lifting by
machine learning technology. Viewers are also often players with copies of the
game already installed on their home systems. Some publishers let these viewers
pay to watch a championship through their own game, driving the camera wher-
ever they like across the playing field, following their favorite champions, and
streaming their views to their own audience. Multiplayer games and the networks
that sustain them are now capable of handling both large numbers of players and
a global audience of social spectators. This rich technological ecosystem has cre-
ated numerous social and economic incentives for stakeholders.
Digital technologies helped a lot the diffusion of sport event content, even not
simultaneously when the event takes place. For fans, it is possible to use many
different technological appliances to play with the match images and the ath-
letes, even in a recorded broadcast. The real sport event seems to become just
the origin of a huge entertainment potential for fans and spectators. The com-
munication part of the event is the rest. Indeed, the sport clubs must be cautious
not to over-promise in terms of how good the game will be or how well specific
athletes may perform. Moreover, sport clubs have no direct control over sport
performance, which is basically perishable. Its perishability forces sport event
organizers to place a great emphasis on the quality of the game experience, which
is more and more ensured, pushing an overall effort on the product extensions
(Mullin et al., 2007; Zagnoli & Radicchi, 2011). To overcome and to diversify
the financial risk connected with the unpredictability of sport performance in
 Disruptive Technologies 59

the contemporary context, sport organizations invest in additional products and

services, collateral to the sport practice itself such as merchandising, selling mem-
orabilia, multi-functional facilities, broadcasting rights, and, last but not least,
eSport entertainment.
Today, the cutting-edge trend related to the product extension strategy is offer-
ing a wide array of services and activities that can be “played” into the stadium or
at home during a broadcast match. The disruptive dimension of this trend is the
amplification of the “winning” and “losing” concepts. Through social, virtual, and
augmented media, while players and athletes compose the real performance on
the pitch, fans and spectators are entertained with virtual or augmented images.
They are even kept busy with social network interactions: building, involving,
and engaging spectators. It is an extension of the product event that enhances
entertainment and even reduces bitterness and disappointment in case of defeat
of fans’ beloved teams. Therefore, the entertainment of digital communication
has a decisive and overwhelming influence on the content and fans’ behavior,
especially in the case of losing performance. As shown in Figure 3.1, digital com-
munication in sport plays a disruptive role compared to the classic relationships
in the marketing mix leverage. Nonetheless, the key point is the product. In this
case, it is the “product-event” that is becoming more and more like a stage in
which not just the main actors, such as real athletes, are playing, but even
e-elements and appliances keep both the fan physically participating in the arena
and thousands of mediated spectators involved.

Different Typologies of Digital Communication

The sport event has, in essence, two distinct intrinsic dimensions: sport supplying
the “practice” and the fruition enjoyed by fans and spectators. In early times, the
participation in the sport event as a practitioner or a spectator was exclusively
live. It was just a direct and real experience. With the advent in the ‘50s of ana-
logical television and with the emergence at the end of the ‘90s of new media, the
sport experience assumed connotations more and more imbued with a mediated
reality. The entry of new technologies within the sport event began in the ‘60s
with the first satellite transmission of the 1964 Tokyo Olympics. Moreover, since
the Japanese Olympic Games, it enriched the participation of both practition-
ers and spectators with multimedia and interactive dimensions by allowing them
to live not just a mediate but even an interactive experience with the event’s
environment and other consumers. Over the years, real sports events have been
increasingly distributed through electronic, multimedia, interactive, and virtual
channels enabled by the emergence of different technological paradigms (see
Figure 3.2). The fans’ experience at games continues to be the highest priority
for sport teams. Nevertheless, “it’s no longer a matter of going to a sporting event
with a hot dog and a beer and taking three and a half hours and driving home
in the station wagon” (Don Garber, MLS Commissioner, 2018). Full adventur-
ous enjoyment of sports still sees many people preferring to go to the stadium.
60 P. Zagnoli and E. Radicchi

Undoubtedly, watching sports live at the stadium creates an intense excitement

in the crowd (Kanosue et al., 2015), deriving from the share of the same emo-
tional environment with thousands of fans cheering the team they support – a
mix of emotions that last longer in the memories of life. In the meantime, not
passionate, especially less identified, fans may find sporting events sleepy, dull,
displeasing, and perhaps distressing.
Sport events are appealing and enjoyable for an increasing number of people,
even with a lack of identification with a team or a sport player. Even the hard-
core fans sometimes experience this feeling. Pivotal Leagues and teams intensify
to enhance fan excitement and enjoyment in a variety of ways: from pre-game
music and video to sophisticated lights systems, electronic signage, and monitor
displays around the venue. The process has established new rules in the Northern
American context and diffusively in the European sporting emerging environ-
ment. In these countries, professional teams and leagues have invested tremen-
dously to enrich the fan’s experience at the game’s venues. The main objective
of the clubs is to offer intense live participation to fans by increasing their overall
comfort and providing for supporters and partners spectators an unforgettable
emotional experience (Pine & Gilmore, 2011).
The stadium, mainly through its architectural design, is not a mere place to play
the game, but it becomes the fulcrum of promotional and marketing activity by
offering a complex mixture of additional services suitable for fans and supporters.
The sport venue is progressively perceived as a theater where athletes play their
performance. Installation of VIP areas equipped with meeting rooms, restaurants,
catering capabilities, etc., besides the so-called Sky boxes. For fans, the experi-
ence is even higher: the sport pitch itself is furnished with enormous screens and
glass-walled rooms that transmit images, virtual and real, and in some stadiums,
like those of NFL American football, acoustic consultants are brought on board
to guarantee maximum amplification of crowd noise, ensuring a “wall of sound”.
Digital technology has a huge impact on how sports buildings are imagined
and designed. There are opportunities for layering the viewing experience – from
seeing different camera angles in the seat to tracking the heart rate, speed, and
impact of players wearing smart clothing on the pitch. With the rise of smart-
phones, mobile devices and social media have become more and more entwined.
As such, sports fans like to be connected to their favorite social media outlets
24 hours a day and are increasingly participating at live sport events using their
phones – ready to tweet, share, like, and take photos. Sports fans attend live
sports games and use a mobile device either before, during, or after the event.
One of the most common activities among fans using smartphones during live
matches is calling, texting, social networking while the actual game occurs. To
increase fans’ virtual participation during the events, some professional clubs like
Liverpool FC, Barcelona FC, Miami Dolphins, San Francisco 49ers have invested
in developing the so-called Connected Stadium. This venue concept sees sporting
arenas improved technically and digitally to ensure that the audience is encour-
aged to interact online from within the stadium. Nevertheless, we are just at the
 Disruptive Technologies 61

beginning of the process. Indeed, social media applications within sport venues
radically change the live event experience. It is a sort of meta-event, a “flagship”
innovation that, on the one hand, is still in an experimental phase; on the other
hand, it requires a huge amount of investment. Indeed, only major Leagues and
clubs that have sophisticated and valuable partners are able to pioneer in this
direction. This, in turn, explains why only a few clubs have been bringing their
arenas into the digital and constant connection.
The new trend is clearly the increasing convergence of the physical and digital
experience allowed by the connectivity placed in the stadium to bring fans closer
to the players. This process is boosted by the application of VR. Overcoming the
too early failed introduction during the 1990s, VR in the contemporary envi-
ronment is positioned successfully in the technological framework. VR made its
reappearance even in sports events, enabling watching the game with headsets
such as Samsung Gear VR, Google Cardboard, and Oculus Rift. With the leap
that VR has taken recently, the sport experience is being amplified and becoming
deeply immersive. Searching the VR and AR applications in the sport sector, we
can enlighten the following trends:

1 The fan can view the action from the player’s perspective in the stadium/
arena using an on-player camera feed. Moreover, the visor enables to have
interactivity, stats, and additional info added to the display. For example, in
2018, player perspective video feeds at several sporting events were offered,
including a Euro league basketball match. The garment contains an embed-
ded HD camera, a microphone, and additional sensors that monitor player
health stats.
2 A VR headset and an app enable the fan to have a look around the stadium
as the action unfolds, all without leaving the comfort of his home. This is the
most realistic experience from viewing a sporting event without attending
it in person. The Rio 2016 Summer Olympic Games kicked off with a solid
VR coverage by broadcasting networks such as NBC and BBC– a first in
Olympics broadcasting – viewable with VR headsets through their respec-
tive apps.
3 The National Basketball Association (NBA) has made the most progress in
adopting VR, but other leagues are not far behind, and VR has already cov-
ered three mid-season National Football League games plus the Super Bowl,
the live-stream broadcast of the International Champions Cup (ICC) soccer
games, a NASCAR (National Association for Stock Car Auto Racing) race,
and a couple of National Hockey League (NHL) games.
4 Since one of the strongest arguments against the use of VR in consuming
sports content is the fact that it takes away the social experience embedded
in the direct attendance – losing part of the pleasure of watching a game with
family and friends – social media companies are moving forward enforcing
the innovation diffusion, trying to combine the virtual experience with the
social connection. For example, some high-tech companies are working on
62 P. Zagnoli and E. Radicchi

displaying – through a visor – the virtual reconstruction of the stadium and

the players in near-real time. Fans are invited to step in and view the envi-
ronment from any viewpoint they want. However, fans appear as avatars and
can interact with each other through VoIP.
5 With AR, the fans can catch visual contents enriched with useful informa-
tion about sports, players, and teams. AR technology can bridge this gap
since the real-world view is supplemented with computer-generated digital
content such as images, videos, or animations that enhance the fan’s per-
ception in real time. For example, tennis fans can follow the players through
the “Hawk-Eye”, a virtual ball tracking system, while challenging line-calling
decisions. In cricket, it is used to depict a batsman’s wagon wheel or a bowler’s
beehive. This system enables to visually track the trajectory of the ball and
display a profile of its statistically most likely path as a moving image. The
most popular example of AR application in sport is probably Pokémon Go.
Using a fan’s smartphone camera and GPS signal, the game makes it seem
as if wild Pokémon are gathering up on the streets of the real world. When
walking around and exploring, fans are alerted on their phones with bits of
grass, which signal a Pokémon’s presence.
6 Many teams have recently implanted a new way to engage supporters during
the matches. For example, Manchester City FC launched in 2016 a game
whose aim was to catch virtual animated Pokémon, which have been super-
imposed, into real-world locations by using a smartphone, both inside and
outside the Etihad stadium. This AR game-based is an additional tool to
enhance fans’ loyalty, especially millennial supporters. On the other side, it
enables the club to raise its mobile revenue charts. Although watching sports
at home does not provide the mediate spectator the same experience as the
fan gets in the stadium, nevertheless, it enables the user to enjoy the sporting
events with a lot of commentaries, discussion, and graphical analysis.
7 The digital broadcasting of sport games is growing its level of involvement
through the utilization of AR. It enables, for example, to see a replay of the
match on the studio floor in the form of 3D animations, to watch the foot-
age of the live game projected in 3D, to visualize the trajectories of the ball
(or player) with the help of virtual lines and curves. Major sports broad-
casters like Fox Sports, Sky Sports, Star Sports, Entertainment & Sports
Programming Network (ESPN), etc., are already using this technology dur-
ing the live broadcasting of sports events. Undoubtedly, the integration of
AR into the broadcasting of sporting events creates a strong immersive view-
ing experience. Ultimately, it improves the engagement of the fans that, by
getting used to this new fruition, are progressively becoming eager of their
“distribution routine” granted by the sport broadcasting network that pro-
vides access to their team.
8 New technology allows offering new products. An example of pure virtual
sport events is fantasy games, one of the most popular activities nowadays
 Disruptive Technologies 63

in the sport industry. While fantasy sports were born of paper, pencil, and
calculator, the preferred venue to trade players is online. Fantasy sports deal
with selecting and forming virtual teams in which each virtual player repre-
sents a real player from a team of the professional league. Based on the actual
performance of the real player from the actual game, the score for each vir-
tual team is decided. Furthermore, the team whose players score maximum
becomes a winner of the league. Thus, the game here is to draft, trade, and
remove virtual players in line with the results of the real sport. CBS, ESPN,
NFL, and Yahoo Sports are among the top Fantasy League Organizers. They
help the fan organize (draft, trade, and remove) his team members through
their website. A start-up firm named “AR Sports” came up with their pat-
ented AR platform for organizing the fantasy sports leagues. This AR Sports
system can be used either as a stand-alone application or by integrating with
the existing host sites. Additionally, this platform enables advertisers to dis-
play interactive advertisements using AR.
9 Even viewer habits are shifting. In 2017, pay TV subscriptions fell by
3.6 million as cord-cutting viewers continued to move to streaming and social
platforms (Nielsen, 2017). Despite these new distribution channels, provid-
ers are generally “trapped” with their traditional business model: gather-
ing audiences and selling third parties’ access to their attention. Facebook
has purchased exclusive rights to livestream tournaments of four popular
eSports organizations, including, for example, “Valve’s Counter-Strike:
Global Offensive Pro League”. Amazon’s streaming service, Twitch, has
acquired exclusive rights to stream Blizzard’s Overwatch League games.
One challenge for the industry – both game publishers and the companies
building and maintaining the infrastructure making multiplayer gaming
possible – has been metrics since participants do not necessarily fit into
traditional TV-industry target markets.
10 The use of data analytics in eSport is evolving as well. For instance, after
Nielsen launched an eSports division, the organization partnered with
Activision – a US-based video games producer – to bring audience analytics
into Call of Duty, one of the most popular multiplayer video games. More
insight into use can help bring familiarity and clarity to the eSports market
while driving advertising revenue to game publishers. The stakes are high: at
least 30 major-player eSports tournaments worldwide are scheduled in 2018,
with top players signed to multi-year contracts worth up to $320,000 annually
(Nielsen, 2017). Success in professional eSports demands rigorous teamwork
and persistent commitment to high performance; winning teams can earn
millions in prize money, with popular teams and players drawing significant
income from advertising, merchandise, and royalties. It is no surprise that
many teenagers seriously aspire to be professional gamers. Some parents are
now taking their children to live eSports events, sharing an experience that
lies at the heart of spectator sports.
64 P. Zagnoli and E. Radicchi

Marketing Implications
The increasing and tangible disruptive process that we tried to document in this
work is mainly diffused in professional mass sport such as the most popular and
well-known sports and teams (soccer, basket, etc.) (Hilvoorde, 2013; Zagnoli &
Radicchi, 2016). The contemporary sport event can really depend on the skill lev-
els of the athletes, the location of the event, the amateur or professional degree of
the team, the technological infrastructure diffusion in a country, the sportive and
financial-economic strength of the sport companies or associations, etc. (Parent
& Smith-Swan, 2012). In Figure 3.3, we try to express the tangible differences
in the feature of the sport event, looking at the traditional elements of the real
sport event and at the observable new way induced by eSport affirmation. The
technological side takes the stage with the mediate and virtual dimensions able
to catch enormous attention by living and mediating spectators during and after
the physical sport event.
Empirical research allows highlighting a wide range of experiential sports
events. The contemporary context sees emerging a great variety of “hybrids”, such
as real events enriched with multimedia and virtual dimensions. Nevertheless,
traditional ways of accessing sport contents such as live participation in terms
of an amazing number of real spectators worldwide are not surpassed by new
technologies. Rather, it seems that interactivity and virtuality are emerging and
co-exist with the real experience, completing the latter with additional multi-
media services. The real dimension feeds the virtual one: almost all the events
involving physical and direct participation can be enriched through a “virtual
world”. New digital technologies appear to play a role in the virtual socialization

Figure 3.3 Features of Sports Events: From Real Events to eSport.

 Disruptive Technologies 65

of people during the event by encouraging group interaction, involvement, and

passion with which fans, sport clubs, and companies must cope with.
Fans form a kind of “community” during the real event, sharing their interest
for specific passions like a team and creating collective identities, which are more
and more “designed” by a gamified experience. Digitalization and virtuality shape
a sort of “meta-event”: fans are immersed in a virtual experience that is lived
simultaneously when the real sport event is taking place. Virtual communities
can enrich the content of the real event. Today, blogs, social networks, and online
communities enable users to live a different experience putting in a virtual envi-
ronment activity, such as commenting with friends a match result or an athlete’s
performance, that before was exclusively done in the real world, especially read-
ing and discussing in circles, bars, pubs, etc., statistics and figures appearing in
dedicated newspapers. By moving from real to virtual events, fans’ engagement
level seems to grow. New technologies allow supporters to express their opinions
by giving voice to their participation and showing their abilities. Social media
have become the new place in which to diffuse actions and discuss them among
fans. In some cases, they can also co-create the event. Therefore, we found a
change not only regarding the places where people access the event – sports phys-
icals venues become the main stage where spectators live an immersive, social,
and “augmented” experience. Moreover, by using social networks, virtual and
AR apps, eSport, etc., fans can fulfil their needs of participation and discussion
between each other.
For sports clubs, this emerging trend is a key advantage. On the one hand, new
technologies increase the entertainment side of the sport event. On the other
hand, they enable sport organizations to develop information feedback with fans
on which to plan new customized marketing activities aimed to increase pas-
sion, identification, and “selling”. Embracing the opportunities and challenges
in eSports can help sport clubs evolve with their audience and transform to meet
changes brought about by digital disruption. Teams’ executives should understand
that modern storytelling includes an important video game market. While sports
audiences are being drawn to more interactive and immersive narrative experi-
ences, the risk is that physical sport becomes more and more just a viable digital
complement. Through the new digital technologies, viewers may become broad-
casters using their own social media devices during the game, and fans are players
while interacting with eSport games. The eSports industry may offer access to
each of these functionalities and enable the fan to play different spectators’ roles
(players, broadcasters, streamers, etc.).
Companies should understand the nuances of each role before moving into
the eSports arena, whether as sponsors, investors, service providers, etc. The
sport event is increasingly a valuable platform to perform media’s brand and
companies (sponsors) names. Everything tends to follow the procedure of the
stage. Digitalization conveys the sport event into the rules of the artificial set.
On the contrary, the focus that puts in value sport and respects sport sustaina-
bility should be based on capturing the local specificity of the place, designing a
66 P. Zagnoli and E. Radicchi

venue that somehow responds to the fan identity and to the culture of the sport
team. Global companies are a master in this process and invest in huge sport
events’ communication and marketing deals. Global firms are, in fact, the biggest
supporters of sport events. Especially sport-related companies – such as sporting
goods suppliers, casual and sportswear producers, retailers, etc. – invest in sports
events addressing the experiential marketing potential. Traditionally sport events
for companies can have multiple connotations: (a) to involve practitioners and
fans, such as prospective fruitful customers, and (b) to engage users in order to
improve product usability and performance. Nevertheless, even non-sport-related
corporations such as manufacturing firms, consumer goods producers (in prime’s
food and beverage, financial, and bank services), and multimedia companies
(Internet, mobile phones providers, online games developers, etc.) are showing
a high “strategic sensitivity” to join, catch, and exploit the people’s emotions in
promoting and selling their products addressed by the sport events.
Obviously, sport contents chosen by companies to enhance their products and
brands’ image are always top-class sport. Such as the most exciting and popular
disciplines (e.g., football), the best global performers teams (e.g., Manchester City
FC, NBA teams, etc.), or the most worldwide appealing major sport events (e.g., the
Olympics, the Football World Cup, the American Sailing Cup, etc.). Industrial
companies’ marketing strategies require jointly promoting their products through
events that represent successful, popular, and spectacular sports, seizing the atten-
tion of wider consumer segments and offering memorable emotions. Companies’
goal is to enhance an identification process of consumers to carry within people’s
daily life the emotions of the sport increasingly driven into preferring their
brands and products. Sport- and not-sport-related firms can partner with teams,
players, Leagues, and broadcasting networks. At the base of these multiple rela-
tions, there are data analytics that is progressively moving into the gaming world.
Social media are an amazing source of data about fans, supporters, and users.
The new media and the sponsor companies have access to a huge amount of
information. Big databases enable identifying users’ behaviors, tastes, preferences,
and characteristics, and profile them according to quantitative and qualitative
statistics and figures remarkably helpful to implement companies’ marketing
strategies. If multimedia companies are strongly impacted using these applications
as tools of sport content distribution and promotion, this phenomenon is quite
important also for other manufacturing firms which use interactive instruments
to appeal to different consumers’ targets by associating their products and brands
with emotional experiences.
Finally, we can say that in the disruption process, digital companies are assum-
ing a central role both in value creation and in sport distribution. Consequently,
sport companies, especially at the global and professional level, must interact and
select the overload of technological possibilities. They should choose among the
mere entertainment technologies the most appropriate for the different sport and
the fan highly differentiated in terms of passion, cultural and economic attitude.
By the way, for sport events, despite enforcing a complex digital value co-creation
 Disruptive Technologies 67

process, team interactions with its fans are still a core dimension. Therefore, in
the rapidly evolving eSport industry, media, entertainment, and professional
sports companies increase access to a valuable global audience, unlock advertising
potential, and develop new entertainment and hospitality offerings, empowering
the main sport franchises. Sport businesses can capitalize on this fundamental
shift in media and sport shaped by social entertainment, live streaming, and
high-intensity competition in the digital world. By highlighting their own passion
for sport and digital gaming, companies can help co-create a more inclusive global
future of interactive and immersive entertainment.
Nevertheless, in the sport sector, media channels have been acquiring progres-
sively more and more power over the content providers (sport clubs, Federations,
Leagues, etc.). As in early times emerged in other industries such as food and large
consumer goods retailing or industrial manufacturing (Lugli, 2007), distribution
channels have a tremendous impact on sports content management. In contrast,
sporting organizations have reduced their power of controlling sport event’s dif-
fusion, taken over by multiple and overwhelming media corporations (Zagnoli &
Radicchi, 2011; 2016), which dictate the characteristics of events and matches,
change fundamental game rules, and influence the ways how fans and spectators
enjoy the sports content. It is crucial for sport clubs to develop a more complex
and structured organization that entails a set of strategic new media capabilities.
Not only to retain the value generated but also to uphold a higher level of control
over their own “identity” content.
Many questions are still open since sport is a socio-cultural and economic com-
plex outcome. How can companies build relationships with players, leagues, and
the millions of spectators watching them? How can broadcast media companies
bring a more enhanced viewing experience to the emerging audience? What are
some of the new merchandising opportunities with downloadable digital con-
tent? Considering that social media, video platforms, and messaging tools have
expanded our physical behaviors into the eSport world, executives across media
and entertainment, sponsor, and sport teams should think carefully about the
involvement and the new centrality of digital communication and the media and
technology that make it possible. Therefore, there are many raising challenges.
However, the main one is how sport practice and competition can protect and
maintain their essence despite sponsorship requirements and digital distribution
imperatives. Will sports be able to keep its identity or become more and more a
marketing opportunity?

References
Buzzetto-Moore, N. A. (2013). Social media and prosumerism. Issues in Informing Sciences
and Information Technology, 10, 67–80.
Deloitte. (2019). ESports graduates to the big Leagues. Center for Technology, Media and
Telecommunications. https://www2.deloitte.com/cn/en/pages/about-deloitte/articles/pr-
deloitte-esports-to-the-big-leagues.html#
68 P. Zagnoli and E. Radicchi

Dosi, G. (1988). Sources, procedures, and microeconomic effects of innovation. Journal of

Economic Literature, 3, 78–99.
Foster, G., Greyser, S. A., & Walsh, B. (2005). The business of sport: Text and cases on strat-
egy and management. Nashville, TN: South-Western College Pub.
Freeman, R. E. (1984). Strategic management: A stakeholder approach. Boston, MA: Pitman.
Freeman, R. E., Harrison, J. S., & Wicks, A. C. (2007). Managing for stakeholders: Survival,
reputation, and success. New Haven, CT: Yale University.
Hester, B. (2018, March 6). Fortnite is currently Twitch’s most-watched game. Rolling
Stone. https://www.pcgamer.com/fortnite-is-now-the-most-watched-game-on-twitch-
doubling-pubgs-viewer-count/
Hilvoorde, I. (2013). Sport and play in a digital world. London, UK: Routledge.
Kanouse, K., Kogiso, K., Oshimi, D., & Harada, M. (Eds.) (2015). Sports management and
sports humanities. London, UK: Springer.
Lugli, G. (2007). Marketing channel. La creazione di valore nella distribuzione specializzata.
Milano, Italy: Utet.
Lusch, R., & Vargo S. (2006). Service dominant logic: Reactions, reflections and refine-
ments. Marketing Theory, 6 (3), 281–288.
Mazmanian, M., & Erikson, I. (2014, April 26–May 1). The product of availability: under-
standing the economic underpinnings of constant connectivity [Conference Paper]. SIGCHI
Conference on Human Factors in Computing Systems. Toronto, Canada. https://doi.
org/10.1145/2556288.2557381
Nielsen. (2017). The eSports playbook: Maximizing investment through understanding the fans.
https://www.nielsen.com/us/en/insights/report/2017/the-esports-playbook-maximizing-
investment-through-understanding-the-fans/#
Normann, R., & Ramirez, R. (1993). From value chain to value constellation: Designing
interactive strategy. Harvard Business Review, 71 (4), 65–77.
Parent, M., & Smith-Swan, S. (2012). Managing major sports events: Theory and practice.
London, UK: Routledge.
Pine, J., & Gilmore, J. H. (2011). The experience economy. Boston, MA: Harvard Business.
Pisano, G. (2006). Profiting from innovation and the intellectual property revolution.
Research Policy, 35 (8), 1122–1130.
Radicchi, E. (2012). Megaeventos deportivos y creación de valor para las economías
anfitrionas. In Llopis Going, R. (Ed.), Megaeventos deportivos. Perspectivas científicas y
estudios de caso (pp. 25–51). Barcelona, Spain: Editorial UOC.
Santomier, J., & Shuart, J. (2008). Sport new media. International Journal of Sport
Management and Marketing, 4 (1), 85–101.
Tappscott, D. (1996). The digital economy: Promise and peril in the age of networked intelli-
gence. New York, NY: McGraw-Hill.
Tappscott, D., Ticoll, D., & Lowy, A. (2000). Digital capital. Harnessing the power of business
webs. Boston, MA: Harvard Business School.
Teece, D. J. (1986). Profiting from technological innovation: Implications for integration,
collaboration, licensing and public policy. Research Policy, 15 (6), 285–305.
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic manage-
ment. Strategic Management Journal, 18 (7), 509–533.
Toffler, A. (1980). The third wave. New York, NY: Bantam.
Vargo, S., & Akaka, M. (2012). Value co-creation and service systems (re)formation: A
service ecosystem view. Service Science, 4(3), 207–217.
 Disruptive Technologies 69

Vargo, S., & Lusch, R. (2011). It’s all B2B … and beyond: Toward a systems perspective of
the market. Industrial Marketing Management, 40 (2), 181–187.
Vargo, S., Maglio, P. P., & Akaka, M. (2008). On value and value co-creation: A service
systems and service logic perspective. European Management Journal, 26 (3), 145–152.
Yoffie, B. (1997). Competing in the age of digital convergence. Boston, MA: Harvard Business
School.
Zagnoli, P., & Cardini, C. (1994). Patterns of International R&D cooperation for new
product development: The Olivetti multimedia product. R&D Management, 24 (1),
37–52.
Zagnoli, P., & Radicchi, E. (2010). The football fan community as a determinant stake-
holder in value co-creation. Sport in Society, 13(10). 1532–1551.
Zagnoli, P., & Radicchi, E. (2011). Sport marketing e nuovi media. Milano, Italy: Franco
Angeli.
Zagnoli, P., & Radicchi, E. (2016). Sport marketing and new media: Value co-creation and
intertype competition. In M. Desbordes, S. Chadwick, & N. Chanavat (Eds.), Routledge
handbook of sports marketing (pp. 277–298). London, UK: Routledge.
Zagnoli, P., & Radicchi, E. (2017, 15–16 June). Sport: The new social media value co-
creation. Paper presented at the XXIX Convegno Annuale Sinergie-SIMA, Università
di Napoli Federico II, Italy.
Zagnoli, P., & Radicchi, E. (2018). The ACF Fiorentina marketing: The strength of local
identity. In M. Desbordes, N. Lorgnier, & N. Chanavat (Eds.), Routledge handbook of
football marketing (pp. 351–371). London, UK: Routledge.
Chapter 4

Challenges in Sport Marketing

A Systematic Review of Digital Sport Marketing
Salvador Angosto, Alejandro Lara-Bocanegra,
M. Rocío Bohórquez, Jerónimo García-Fernández,
and James J. Zhang

Introduction
“The change has come. The change is present”. These two phrases could be a pre-
view of the changes that are being promoted in organizations around the world,
possibly due to the digital transformation that is taking place. The imperative
need to work in parallel with the face-to-face and digital has had an impact on
the opening of a new door not only to be explored but also to be implemented.
Thus, digital transformation refers to new digital business models, a digital
consumer, collaborative work, the implementation of analysis data, continuous
innovation, the implementation of digital strategies, the change of organiza-
tional culture, creation and social media management, critical thinking, and of
course, the attitude and aptitude of the people who make up the organizations.
Actually, it refers to the permanent change of organizations through technologies
since an improvement will always appear in the last update. That is why digital
transformation is not the use of the latest technologies but the correct integra-
tion of them in organizations allowing the evolution of organizations and their
development (Kane et al., 2015). The transformation could also be encouraged
by changes in consumption. Precisely the ways of relating and communicating
between organizations and consumers have changed. Now, in addition to “face-
to-face” relationships, organizations are seeking to create relationships with their
users and consumers in digital format (Phillips, 2015). In fact, most organizations
are using different tools and have created profiles on social networks for greater
contact and a greater relationship with their current and future consumers. This
new perspective would therefore be focused on the new forms of marketing that
organizations use and that ultimately are based on digital marketing.
Kannan and Hongshuang (2017) indicate that digital marketing may be seen
as activities, institutions, and processes facilitated by digital technologies for cre-
ating, communicating, and delivering value for customers and other stakeholders.
Specifically, they define it as “an adaptive, technology-enabled process by which firms
collaborate with customers and partners to jointly create, communicate, deliver, and
sustain value for all stakeholders” (Kannan & Hongshuang, 2017, p. 23). In fact,
these new forms of communication create value through digital environments,

DOI: 10.4324/9781003302476-4
 Challenges in Sport Marketing 71

creating new online experiences for the consumer. In this new form of communi-
cation, digital marketing must take into account the environment in which the
organization operates to make an analysis of consumer behavior, search engines,
contextual interactions, or possible “digital” competitors. These aspects are nec-
essary to understand and analyze in order to establish actions with which the
objectives of the marketing plan are achieved. Likewise, understanding the mar-
ket that surrounds the organization and, in turn, specific to the sector helps to
establish objective goals to be achieved through digital sport marketing.
Among the most important activities that are carried out, digital marketing is
based on a strategy to be developed, focusing firstly on capturing the attention of
the organization, the second is converting those contacts into potential clients,
and a final phase that would analyze whether the measures and actions taken are
consistent with the needs of consumers. In fact, digital marketing has changed
the way of attracting customers and keeping them over time.
Digital marketing uses the website, email, apps, etc., and of course, social media.
Precisely in the sports sector, social media have been the most used tools in dig-
ital marketing to reach more consumers and communicate with them (García-
Fernández et al., 2017). In fact, He et al. (2013) stated that organizations use them
to find new customers, improve user satisfaction, increase sales and revenue, retain
them and consequently build a good reputation for brand image. For this reason,
social media are the perfect tool to establish relationships with customers or poten-
tial customers, favoring repurchase and, therefore, loyalty (Yoshida et al., 2014).
In the sports sector, social media are defined as “new technologies that facilitate
interactivity and co-creation that allow the development and exchange of content
generated by users between organizations (for example, teams, bodies, agencies and
media groups) and individuals (for example, consumers and athletes)” (Filo et al.,
2015, p. 167). Thus, its importance lies in the continuous exchange of information
from the organization to the consumer himself, managing to increase the knowledge
of both parties. In fact, and due to the growing use by users of these platforms, com-
panies from the different business sectors of sport are looking for strategies to make
themselves known, improve their brand image, and reach their target audience
(García-Fernández., 2015). For this reason, the relationship between digital sport
marketing and social media is obvious, taking into account that social media have
revolutionized communication in the sports industry and therefore in how sports
organizations interact with consumers (López-Carril et al., 2020). However, so far,
no study has been carried out on the specific studies that have analyzed digital mar-
keting actions and their tools in sport. Therefore, the aim of this study is to carry
out a systematic review of literature on digital sport marketing in the sports sector.

Method
To conduct the comprehensive review of literature, the search terms were estab-
lished in two groups; one group included terms related to digital marketing: “dig-
ital market*”, “digital communication”, “internet market*”, and “e-marketing”; while
72 S. Angosto et al.

the other group included terms related to sport: “sport*”, “fitness”, and “physical
activit*”. Different databases were selected to include a wide range of areas related
to this interdisciplinary study, including sports science and marketing. The data-
bases used were Web of Science, Scopus, SPORTDiscus (EBSCO), and ABI/
Inform (Ovied). The search was conducted between September 03 and December 18,
2020. The search covered all years, and no language limitations were imposed,
but the type of document was exclusively limited to scientific articles.

Inclusion and Exclusion Criteria

For the purposes of this review, we included empirical papers in peer-reviewed
journals, excluding dissertations and abstracts. Grey literature was not included,
ruling out evaluation reports, annual reports, articles in non-peer-reviewed jour-
nals, and other means of publication. The inclusion criteria for the articles in the
search were the following: (i) journal articles; (ii) publications in English; (iii) use
digital marketing in a sport context; and (iv) articles focused on online advertis-
ing and not on the consumer. As exclusion criteria have been used: (i) Congress
proceedings, book chapters, books, or other types of publications; (ii) no digital
marketing actions; (iii) theoretical studies, qualitative approach or reviews; (iv)
articles in a language other than English, and (v) duplicate articles.

Assessment of Methodological Quality

The Methodological quality was assessed using a 20-item tool adapted by the
authors to the context of sports marketing study typology in which there are no
intervention processes on the subjects of the Consolidated Standards of Reporting
Trials (CONSORT) checklist (Schulz et al., 2010). Each study was independently
scored by two reviewers evaluating the different sections that make up the studies
and scoring each item with 1 if the study satisfactorily met the criterion, and
with 0 if the study did not satisfactorily meet the criterion or if the item was not
applicable to the study. Disagreements between the reviewers were resolved by
checking and discussing the original study until a consensus was reached.

Data Extraction and Synthesis

Figure 4.1 shows the Flow Diagram proposed by Moher et al. (2009) following
the PRISMA methodology in all points that could be common to a systematic
review of these characteristics. The initial database search returned 2,218 results,
reduced to 1,573 once duplicates were eliminated. Two reviewers conducted a full
scan of the title, then an abstract review, and finally a full-text review using the
inclusion and exclusion criteria. The number of final documents for the qualita-
tive synthesis was 13. Among the articles that remained at the abstract level
(n = 4), a third reviewer also examined the abstracts of the articles to confirm
their eligibility, and there were no discrepancies with the two previous reviewers.
 Challenges in Sport Marketing 73

A form was developed for data extraction that included the following aspects: (a)
year of publication; (b) country; (c) journal title; (d) theory; (e) sport; (f) type or
organization; (g) method; (h) data collection; (i) platform; (j) analyses performed;
(k) main results; and (l) applications.
To test quality, the risk of bias analysis of the 14 studies evaluated in the
research showed that no studies had a high score of 15 points or more out of
20 totals. All studies except one had a mean score between 10 and 15 points,
and only one study had a score below 10 points (Ioakimidis, 2010). It should be

Figure 4.1 Flow Diagram. (Source: Moher et al., 2009.)

74 S. Angosto et al.

noted that none of the studies analyzed carried out a calculation of the sampling
required for the generalization of the results, which could be due to the fact that
all the studies carried out a selection of the sample for convenience within a cer-
tain population. No studies indicated the author who carried out each part of the
research or funding.

Results and Discussion

Table 4.1 shows the results about the academic variables of the articles. The pub-
lication period covered from 2004 to 2019, and no articles were found in 2020
that met the criteria. Approximately one half of the studies have been published
in the last three years, with four studies being published in 2019 (Burton, 2019;
Naraine et al., 2019; Tejedor et al., 2019; Zanini et al., 2019). The years 2010 and
2018 had two publications in each case (Baena, 2018; Hazari, 2018; Ioakimidis,
2010; Tsitskari et al., 2010). The country of publication of the first leading author
was the United States with four publications (Dick & Turner, 2007; Filo & Funk,
2005; Hazari, 2018; Popp et al., 2017), followed by Australia with three publica-
tions (Evans & Smith, 2004; Filo et al., 2009; Naraine et al., 2019). Two countries
had two publications, Greece (Ioakimidis, 2010; Tsitskari et al., 2010) and Spain
(Baena, 2018; Tejedor et al., 2019), and finally, with one article Brazil (Zanini
et al., 2019), India (Kumar & Bagchi, 2020), and Canada (Burton, 2019).
All the articles were published in a total of eight journals, mainly in the
sports context (10/14 articles), being three of the non-sport-related journals Social
Responsibility Journal, Future Internet, and Marketing Intelligence & Planning. Five
articles were published in the specialized journal International Journal of Sports
Marketing & Sponsorship (Burton, 2019; Evans & Smith, 2004; Hazari, 2018;
Ioakimidis, 2010; Popp et al., 2017), two articles in Sport Marketing Quarterly
(Dick & Turner, 2007; Filo & Funk, 2005), while the rest of the journals had one
publication. As it can be seen, most of the publications (7/14) have been made in
specialized sport and marketing journals; there is also a publication in the Journal
of Sport Management, a global reference journal in sport management. Only
two studies expressly indicated that they used a previous theory to develop the
research, the Social Identity Theory (Baena, 2018) and the Congruency Theory
(Hazari, 2018).
The sports field on which the research was based showed that seven studies
focused on digital marketing in sports organizations and five studies on sports
events. Filo and Funk (2005) carried out their study in a sports organization and a
sports event. One aspect of the sports on which studies have focused was that soc-
cer was the most representative, with four articles published recently in the last
three years (Baena, 2018; Burton, 2019; Tejedor et al., 2019; Zanini et al., 2019).
Basketball was represented in two studies (Dick & Turner, 2007; Naraine et al.,
2019), a study on motor racing (Filo et al., 2009) and the Olympic Games (Hazari,
2018). Two studies examined professional leagues in Australia, looking at rugby,
Australian football, soccer, and basketball leagues (Evans & Smith, 2004), while
Table 4.1 Academic Variables of the Articles

Authors Year of Publication Country Journal Theory Sport Focus

Baena 2018 Spain Social Responsibility Journal Social Identity Soccer Organization
Theory
Burton 2019 Canada International Journal of Sports Not specified Soccer Sport event
Marketing and Sponsorship
Dick & Turner 2007 United States Sport Marketing Quarterly Not specified Basket Sport event
Evans & Smith 2004 Australia International Journal of Sports Not specified Multiple sport Organization
Marketing and Sponsorship leagues
Filo & Funk 2005 United States Sport Marketing Quarterly Not specified Women’s Sport Organization/
Clubs Sport event
Filo et al. 2009 Australia Journal of Sport Management Not specified Motor race Sport event
Hazari 2018 United States International Journal of Sports Congruity theory Olympic Games Sport event
Marketing and Sponsorship
Ioakimidis 2010 Greece International Journal of Sports Not specified Multiple sport Organization

Challenges in Sport Marketing 75

Marketing and Sponsorship leagues
Kumar & Bagchi 2020 India Annals of Tropical Medicine & Not specified Not specified Sport event
Public Health
Naraine et al. 2019 Australia Communication & Sport Not specified Basket Organization
Popp et al. 2017 United States International Journal of Sports Not-specified Athletics Sport event
Marketing and Sponsorship Department
Tejedor et al. 2019 Spain Future Internet Not specified Soccer Organization
Tsitskari et al. 2010 Greece Serbian Journal of Sports Sciences Not specified Soccer/Basket Organization
Zanini et al. 2019 Brazil Marketing Intelligence & Planning Not specified Soccer Organization
76 S. Angosto et al.

the other study compared teams from US soccer and ice hockey leagues with UK
soccer and rugby leagues (Ioakimidis, 2010). Finally, one study looked at teams
from a women’s sports organization (Filo & Funk, 2005), and another study eval-
uated the sports departments of the NCAA Division I universities (Popp et al.,
2017). Kumar and Bagchi (2020) do not specify the type of sport attended by the
participants in their study.
The results on the methodological aspects of the different studies are shown
in Table 4.2. The objectives that the different studies had can be summarized in
four different approaches. The first and most numerous approach (8/14) has been
the studies that have evaluated or analyzed marketing strategies and techniques
used by organizations and sporting events (Dick & Turner, 2007; Evans & Smith,
2004; Filo & Funk, 2005; Ioakimidis, 2010; Kumar & Bagchi, 2020; Popp et al.,
2017; Tejedor et al., 2019; Tsitskari et al., 2010). The analysis of these marketing
techniques or strategies focused mainly on the actions involving increased sales
of tickets for matches or sports events. The second approach of the studies on
digital marketing was focused on the impact of the organizations on the website
or social networks (5/14), approaching the evaluation of the media possibilities
and possible advertising actions in them associated with the organization or the
interaction with consumers (Filo et al., 2009; Hazari, 2018; Naraine et al., 2019;
Popp et al., 2017; Tsitskari et al., 2010). Third, two studies focused on assessing
consumer commitment to brand love (Baena, 2018; Zanini et al., 2019). Finally,
one study evaluated the feelings of social network consumers toward the market-
ing campaigns of sponsoring brands and the ambush marketing campaigns during
a sporting event (Burton, 2019).
Regarding the method used to develop the research, nine studies used quanti-
tative methods, four studies used mixed methods (Baena, 2018; Filo et al., 2009;
Kumar & Bagchi, 2020; Zanini et al., 2019), and one study used observational
methodology to assess the characteristics of professional team websites based on
a category system (Ioakimidis, 2010). Studies using qualitative methodology also
used interviews, open-ended questions, and tweet content analysis. The remain-
ing studies used a variety of quantitative methodologies such as a telephone ques-
tionnaire (Evans & Smith, 2004), paper-based survey (Dick & Turner, 2007; Filo
& Funk, 2005; Filo et al., 2009; Hazari, 2018; Tsitskari et al., 2010), and online
survey (Baena, 2018; Kumar & Bagchi, 2020). Three studies used a social network
analysis tool (Burton, 2019; Naraine et al., 2019; Zanini et al., 2019), two studies
were based on information available on the website (Pop et al., 2017; Tejedor
et al., 2019), and information from financial databases (Popp et al., 2017).
The platform most used by researchers to carry out their studies has been the
website, being used in a total of eight studies out of 14 analyzed (Baena, 2018;
Dick & Turner, 2007; Evans & Smith, 2004; Filo & Funk, 2005; Filo et al., 2009;
Ioakimidis, 2010; Tejedor et al., 2019; Tsitskari et al., 2010). Five articles evaluated
social networks such as Facebook, or primarily Twitter, as an ideal digital market-
ing tool (Baena, 2018; Burton, 2019; Naraine et al., 2019; Popp et al., 2017; Zanini
et al., 2019), and one study evaluated multiple platforms without specifying which
Table 4.2 Methodological Aspects of the Articles

Authors Objectives Method Data Collection Platform Analysis

Baena (2018) To provide a better understanding Mixed methods Online survey/ Website and Content analysis,
of the effect that the corporate interview social networks regression
social responsibility (CSR)
practices might have on brand love.
To analyses the importance of
making supporters aware and
involved in such initiatives.
Burton (2019) To explore consumers’ attitudes Mixed methods rTweet’s Twitter Twitter Comparative
toward ambush marketing and data
official event sponsorship through
the lens of sentiment analysis.
To examine social media users’
ethical responses to digital event
marketing campaigns during the
2018 FIFA World Cup.
Dick & Turner (2007) To determine whether the Quantitative Survey Website Comparative

Challenges in Sport Marketing 77

marketing techniques that NBA
marketing directors viewed as
valuable and useful were viewed in
a similar fashion by attendees at
NBA games.
Evans & Smith (2004) To identify any gaps between seven Quantitative Telephone- Website Descriptive
empirically derived theories of administered
internet marketing ‘best practice’ survey
and the activities of professional
sports clubs.
To provide recommendations for
bridging any identified gaps.
(Continued)
Table 4.2 Methodological Aspects of the Articles (Continued)

78
Authors Objectives Method Data Collection Platform Analysis

S. Angosto et al.
Filo & Funk (2005) To evaluate specific features of Quantitative Surveys Website Content analysis,
sport organizations that comparative
consumers identify as attractive-
To compare such features to the
virtual content presented as part
of the sport organization’s
Internet marketing
communication.
Filo et al. (2009) To develop a better understanding Mixed methods Survey and open- Website Comparative
of online information ended questions
requirements for sport event
organizations.
To evaluate the impact of Website
marketing communication on
consumer motivation and
attitudes toward the event.
Hazari (2018) To investigate attitude toward Quantitative Survey Multiple Comparative,
sponsorship outcome as it relates regression
to purchase behavior, gender,
sponsor patronage, sports
enthusiasm, and social media
consumption.
Ioakimidis (2010) To identify and discuss opportunities Quantitative Observation Website Descriptive
for a sport organization to increase
its fan base and online user
interaction as a key marketing
strategy.
(Continued)
Table 4.2 Methodological Aspects of the Articles (Continued)

Authors Objectives Method Data Collection Platform Analysis

Kumar & Bagchi (2020) To identify productive digital Mixed methods Survey and open- Streaming Descriptive
strategies of sporting events. ended questions Platforms
To study the impact of various
attributes that contribute in the
success of sporting events through
online consumption.
Naraine et al. (2019) To examine the online brand Quantitative Commercial social Twitter Comparative
community of a professional network analysis
sport organization formed via a platform
hashtag marketing campaign to
highlight:
(a) the types of communication
networks that are formed, and
(b) the types of segments derived
from the hashtag.
Popp et al. (2017) To examine the relationship Quantitative NCAA.org and Facebook/Twitter Descriptive,

Challenges in Sport Marketing 79

between growth in social media Financial Database regression
engagement, as defined by the
annual percentage increase in
Facebook Likes and Twitter
Followers, of US college athletics
departments and outcome
metrics of attendance and ticket
revenue.
To measure the impact of social
media as a marketing tool for
NCAA Division I college athletic
departments.
(Continued)
 80
S. Angosto et al.
Table 4.2 Methodological Aspects of the Articles (Continued)

Authors Objectives Method Data Collection Platform Analysis

Tejedor et al. (2019) To analyze digital communication Quantitative Team website Website Descriptive,
of the main football clubs in comparative
Europe to identify and describe
what strategies they follow to
make themselves known on the
internet and to interact with
their users.
Tsitskari et al. (2010) To create the SIMEvI – Sport Quantitative Survey Website Content analysis,
Internet Marketing Evaluation comparative
Instrument, a valid and reliable tool
with Greek sport teams’ fans.
To evaluate the marketing variables
appearing on the websites and not
the variables dealing with the
design, navigation or speed of the
site.
Zanini et al. (2019) To identify how consumer Mixed methods Tweets Twitter Descriptive
engagement practices shape the
dynamics of a soccer club virtual
brand community.
 Challenges in Sport Marketing 81

ones it used (Hazari, 2018). Kumar and Bagchi (2020) evaluated specific sport
streaming platforms users. The type of analysis used showed that the compara-
tive statistic (8/14) was the most used by the researchers (Burton, 2019; Dick &
Turner, 2007; Filo & Funk, 2005; Filo et al., 2009; Hazari, 2018; Naraine et al.,
2019; Tejedor et al., 2019; Tsitskari et al., 2010), followed by six descriptive studies
(Evans & Smith, 2004; Iaokimidis, 2010; Kumar & Bagchi, 2020; Popp et al., 2017;
Tejedor et al., 2019; Zanini et al., 2019). Three studies conducted content analysis
(Baena, 2018; Filo & Funk, 2005; Tsitskari et al., 2010), and another three studies
conducted regressions (Baena, 2018; Hazari, 2018; Popp et al., 2017).
The results on the main findings and applications that can be obtained from
the studies (Table 4.3) show the multitude of approaches and methodologies used
for their development. Following the line previously exposed about the group-
ing of objectives of the studies, those investigations related to the techniques
and strategies of marketing, Dick and Turner (2007) found differences of opinion
between Ticket Holders and Marketing Managers, evaluating better the second
one, about the technique of marketing focused on the offer via email and website.
On the other hand, Evans and Smith (2004) found that, in product marketing,
providing useful and timely information by expanding the brand allows for the
generation of more profits. In addition, these authors specify that an appropriate
strategy for promoting digital marketing is to generate web alliances with other
organizations. Filo and Funk (2005) found congruence between the characteris-
tics and information provided on the website about the products from the viewer’s
perspective; however, there is great variability in the presentation and accessi-
bility to the products, and it is necessary to develop content guidelines for the
leagues. On the other hand, Kumar and Bagchi (2020) showed that 77% of the
participants considered that digital marketing helps to increase the live events
promotion, while 81% considered that it allows to generate more revenues. Also,
they found that two-thirds of the participants consider that if they are willing to
consume more sports events in streaming rather than live.
Ioakimidis (2010) compared American and British leagues and found that the
American leagues offered a greater number of services, with all the teams using
online sales through the website as a marketing tool. Another study focusing on
the NCAA Division I university sports departments concluded that the use of
social media was not a predictor of purchase and attendance at the event, as were
other factors related to team performance, program history, or conference affilia-
tion (Popp et al., 2017). In contrast, Tejedor et al. (2019) found that communica-
tion management is effective for sports teams, yet it did not make good use of all
the possibilities that the digital arena allows. The majority of clubs incorporated
online shops on their websites and included advertising within the multimedia
content. The tweets that contained ticket sales promotion for the sports teams’
matches had a greater number of user interactions than the tweets that presented
sponsorship content (Zanini et al., 2019). The elements of the marketing mix
have also been found to be strong predictors in the evaluation of digital sport
Table 4.3 Main Applications of the Studies

82 S. Angosto et al.
Authors Main Results Applications

Baena (2018) There is a positive and significant association between The model developed can be useful in increasing knowledge
the use of social networks (Facebook and Twitter) to about brand love in marketing.
update fans about the CSR actions carried out by the This study allows sports managers to use the results as a
club with brand love. strategic guide for future plans as well as expand and
reinforce their fan base.
Social practices create a strong link between fans and the
sports brand.
Burton (2019) User attitudes toward ambush marketing appear to The commitment of the sponsors had an illustrative
be significantly more positive than previously assumed, competitive advantage for the brands through the official
as social media users were very receptive to non- sponsorship.
sponsored creative and value-added campaigns. Sponsor brands should be aware of online consumer usage
The results suggest that event partners still have a trends for promotional hashtags and should adopt evaluative
considerable advantage by not maintaining consumer social metrics accordingly.
attention and engagement. Ambush marketing takes an optimistic view of consumer
Non-sponsored campaigns were well received by sentiment, which depends on the delivery and effectiveness of
users at the launch of their campaigns and during the the partnership created, and the added value for consumers.
early stages of the event, but commitment decreased Sentiment analysis offers professionals the opportunity for
significantly over the course of the event. direct interaction and meaningful engagement, but it must
also be representative of the brand’s objectives and approach.
Dick & Turner (2007) There were significant differences between ticket Not specified
holders’ and marketing managers’ perceptions of the
importance of the Internet e-mail offer and the
website.
Ticket holders rated this marketing technique as the
eighth most important, while marketing managers
rated it as the sixth most important.
(Continued)
Table 4.3 Main Applications of the Studies (Continued)

Authors Main Results Applications

Evans & Smith (2004) The most predominant goals of the website were to Information is essential for the development of appropriate
provide useful and timely information, expand Internet strategies so that the scope of the Internet
existing brand information, and generate revenue. marketing opportunity and practical recommendations are
Two-thirds consider that websites enable the achieved in an appropriate manner.
generation of profit, and 58.33% indicate that The review of goals is necessary for the development of
profitability is a goal. leadership and perceptions of e-commerce, linked to the
The most widely used strategic Internet marketing strategic goals to gain competitive advantages in marketing.
tools are those focusing on partnerships/alliances It is interesting to generate online forums that improve the
on websites and link-based or individually managed position of organizations on the Internet and allow them to
websites. form partnerships or alliances.
A large and loyal online community generates substantial
advantages and income associated with its products and the
creation of attractive spaces for sponsors.
Filo & Funk (2005) There is some congruence between the product The study shows a complementary process for effective

Challenges in Sport Marketing 83

characteristics identified by viewers and the research and development in an integrated Internet market.
product communication on the website. The links should examine the content guidelines of each
The variability in the presentation and accessibility team’s website.
of these factors suggests the need for league-wide The degree and rigor of Internet guidelines that are
content guidelines, and efforts should also be established will vary according to the stage of the life cycle in
intensified to ensure that consumer interest which the product is found and the degree of success of the
helps to better shape Internet marketing activities. league itself.
Product presentation and accessibility were different
from the other variables in the marketing mix.
(Continued)
Table 4.3 Main Applications of the Studies (Continued)

84 S. Angosto et al.
Authors Main Results Applications

Filo et al. (2009) The consumer’s experience on the website is more The 15 themes identified provide a useful template for sports
satisfying when they participate in the targeted events organizations to ensure that appropriate information
retrieval of information, rather than simply is available to consumers through their website, the websites
participating in the acquisition of incidental they can connect to, or other communication channels.
information through the multimedia aspects of the Visiting websites and retrieving targeted information led to a
website. significant increase in the beliefs and feelings of those with
Providing the 15 specific topics of information low motivation.
required through website communication can create Sports event organizations can immediately capitalize on
favorable attitudes toward the event among changes in consumer behavior intent by facilitating the next
consumers with low motivation to attend. stage, the purchase of tickets for the event.
Hazari (2018) Social media consumption is positively related to the Provide an opportunity for merchants to take advantage of
attitude toward the event and sports sponsorship. social media networks for sponsorship communication.
Marketing to a target audience of a specific genre thatThe use of sponsorship as a marketing communication tool
plays competitive sports and is active in social media can influence the constructs used in the study, such as
would further impact the results of sponsorship. attitude, sports sponsorship, and purchase intent.
The purchase forecasts as a result of seeing an Companies can use celebrities in sponsorship messages, which
advertisement in social media are gender, competitive can help make the product or company brand more relevant.
sport practice and social media consumption. New communication channels, such as social media, have the
potential to amplify the sponsors’ message due to the
networking effect of users sharing content on the social
network.
Ioakimidis (2010) All teams established a website and offer online media Professional teams could benefit from studying competitors’
services and interaction activities for fans. online practices, as access to the team’s website is not restricted.
The American league’s offer more services than the Teams with more services tend to be more complex to navigate
English ones. than those offering fewer services.To reduce the difficulty of
All teams used the sale of merchandise as a marketing navigating the website, a team’s website could be linked to one
tool on the website. or more secondary sites that focus on specific services such as
Nine teams offer mobile services. the offer for younger fans.
(Continued)
Table 4.3 Main Applications of the Studies (Continued)

Authors Main Results Applications

Kumar & Bagchi (2020) About 77% of the participants consider that DM has The entrepreneur or sports marketer needs to understand
increased the promotion of live events, while 81% digital marketing strategies to initiate digital campaigns that
consider that DM allows them to generate more have a maximum reach.
revenues. The global trend is that different digital platforms experience
71% consider that DM has changed sports large traffic of consumers looking for information,
consumption since it allows them to get to know entertainment, etc., so a huge task is to predict the right
other non-existential sports through digital framework to apply digital marketing.
platforms, especially social media.
Two-thirds of the participants consider that if they are
willing to consume more sports events in streaming
rather than live.
Naraine et al. (2019) The Toronto Raptors’ #WeTheNorth campaign has There is an ongoing research arm focused on the top-level
produced a tightly knit community of online users. implications of social media and sport, in particular, the
Groups play an important role in facilitating the content, narrative, and text produced by the brand’s
empowerment of other sub-groups. consumer.
The brand’s branded anchor, designed to solicit The analysis indicates that the points where the hashtag

Challenges in Sport Marketing 85

“Raptor fans”, has also facilitated the development of network experiences a marked increase in the number of
other sub-groups (followers) through the digital tweets and users interacting are attributable to important
junction itself. time points for the team.
The bias toward male fans and the lack of female fans
demonstrate an opportunity for future growth.
Promotional contests in which fans have to mention or tag
their friends and use the denoted hashtag can have an impact
on the frequency of conversations but also on the density of
the network, which can cultivate fandom and lead generation.
An alternative strategy could be to focus on existing fan
segments themselves and seek to develop content associated
with those segments.
(Continued)
 86
S. Angosto et al.
Table 4.3 Main Applications of the Studies (Continued)

Authors Main Results Applications

Popp et al. (2017) The social media measures were not statistically The findings suggest that increased interactions with social
significant predictors of attendance or ticket media may not necessarily achieve the marketing objectives
revenue. related to increased attendance or income.
The regression models were highly predictive Sports marketers will refine and improve the way they use
(between 53% and 88%), and most of the explained social media as a marketing tool in the coming years.
variations in attendance and ticket revenue were Sports marketers directly attribute ticket sales to social
attributed to team performance, program history, media marketing efforts, often describing sales that are
and conference membership. likely to occur even without social media marketing.
Social media marketing can be much more effective in
achieving other marketing objectives such as branding,
relationship building, and gathering market research.
Tejedor et al. (2019) Communication management is effective, but it also Communication and digital presence on the Internet can be
warns that none of the equipment analyzed takes improved by placing more emphasis on interaction, as this is
full advantage of the possibilities for user interaction a critical aspect of achieving user-level navigation.
offered by the digital scenario. The 2.0 and 3.0 environment offers many possibilities yet to
The clubs offer on average six different languages on be discovered.
their websites. It would be interesting to encourage research on aspects
All but one of the clubs have a shop on the menu, such as transmedia narrative or storytelling, which could
and only three clubs have e-commerce. enrich digital content and improve the user experience.
Ten clubs use the commercial domain. It would be interesting to encourage the interaction between
Nine clubs have advertising on their multimedia and the follower and the club through the collaboration
interactive resources. platforms and the digital communication tools that
cyberspace offers.
(Continued)
Table 4.3 Main Applications of the Studies (Continued)

Authors Main Results Applications

Tsitskari et al. (2010) The main elements of the sports marketing mix Teams can gain a competitive advantage for their customers
(product, price, promotion, public relations) were the through successful e-marketing of their websites; they can
ones that most affected the public when evaluating experience significant improvements in many aspects, such as
web marketing. their fan databases, financial gains, innovations, or business
The first factor was “Product”, which represented process efficiency.
23.2% of the total variation, the second factor,
“Price – place” was developed from the variables
related to online sales, represented 17.4%, the
third-factor “Promotion” explained 13.5%, and the
fourth factor “Public Relations” explained 9.5%.
Males obtained higher scores than females.
Zanini et al. (2019) The dynamics of the engagement is based on two This study offers the refinement of social media strategies and

Challenges in Sport Marketing 87

types of practices: those involving tweeting, the development of content to make them more efficient and
retweeting, answering, mentioning, and liking to establish a relevant communication channel with
messages from and about the profile of the São audiences.
Paulo FC football club. This research provides an overview of social media marketing
The entries related to tweets averaged 106.7 retweets, efforts to build brands and better connections with
296 likes, and 402.8 interactions. The sponsor consumers, considering three levels of contribution as
had an average of 64.3 retweets, 227.1 likes, and follows: BMI strategy, social media strategy, and digital
291.3 interactions. marketing metrics.
88 S. Angosto et al.

marketing, highlighting the product over the other elements; in addition, they
affect differently according to gender (Tsitskari et al., 2010).
The consumer experience will be more satisfactory when consumers are
involved in the generation of information if they participate only in the acqui-
sition of such information (Filo et al., 2009). These authors identified 15 specific
information topics on the website that allow for the creation of more favorable
attitudes toward a sports event among consumers with low motivation. In the
same line of sports events, social media consumption was positively related to a
higher attitude toward sponsorship at the event, as key elements in social media
for purchase, gender, competitive sports practice, and consumption at the event
(Hazari, 2018). Burton (2019) observed that ambush marketing campaigns during
a sporting event showed positive feelings among social media users, especially in
the early and late stages of the sporting event, while sponsorship campaigns dur-
ing the event gained greater competitive advantage and consumer attention. The
campaigns started on Twitter through specific hashtags produced a more united
community within the sports organization of its online consumers (Naraine
et al., 2019). The different groups that exist around the community itself promote
the empowerment of other subgroups through the digital union of users. Baena
(2018) found a positive association between the use of social networks and the
organization’s corporate social responsibility actions led by brand love. Finally,
the dynamics of the engagement were based on those that comprise the actions
of interaction in social media and on the profile of the sports club (Zanini et al.,
2019).
All studies except Dyck and Turner (2007) presented practical applications
based on their results. Thus, the main applications stated by the different authors
have been based on recommendations or guidelines to improve digital communi-
cation on the internet, such as the studies carried out by Baena (2018), Filo and
Funk (2005), and Filo et al. (2009). Baena (2018) specified that social practices
could create a strong link between brand love and fans, so sports managers can
apply these strategies with the aim of increasing the fan base. The degree and rigor
of website content guidelines imposed by sports organizations will vary depending
on the life stage of the product or the success of the organization itself (Filo &
Funk, 2005). Filo et al. (2009) identified 15 essential topics that a sports event
should report on its website. In addition, sports event organizers should quickly
capitalize on changes in consumer behavior intent to facilitate the purchase of
tickets for the event itself. In contrast, Ioakimidis (2010) found that teams with
more services have websites that are more difficult to access, so they should link
services to other secondary sites that can focus on offering services to young fans.
The global trend in the different digital platforms experience indicated that there
is a large traffic of consumers looking for information or entertainment, which
allows sport marketers to develop appropriate product promotion campaigns
(Kumar & Bagchi, 2020).
The studies that have focused on its practical applications oriented toward
improving communication and interaction with consumers highlight the fact
 Challenges in Sport Marketing 89

that information is fundamental for the development of strategies on the Internet

(Evans & Smith, 2004). The 2.0 and 3.0 environments offer many possibilities
still to be discovered by marketing agents, since communication and digital pres-
ence on the Internet can be improved by putting more effort into interactions
with consumers (Tejedor et al., 2019). It is currently known that the different
sports organizations use promotional content as a stratagem to encourage the
consumers’ participation in online platforms (Naraine & Parent, 2017; Stavros
et al., 2014), which will depend on the size of the organization and the context
(Gómez et al., 2019). Popp et al. (2017) suggest that increased social media inter-
action may not necessarily achieve the marketing objectives related to improving
sports event attendance and ticket sales. Instead, this type of marketing can be
effective in achieving other objectives such as building brands, establishing new
relationships, or collecting market research information. In fact, sponsorship is
considered very useful in achieving brand objectives (Johnston & Spais, 2016).
Zanini et al. (2019) offer a vision of social media marketing to promote brands
and generate connections with the consumers through improved the marketing
strategies and the development of efficient content. In turn, sports organizations
and teams can gain competitive advantage for consumers through digital web-
site marketing, which can significantly improve fan databases, financial benefits,
innovation, and efficiency in business processes (Tsitskari et al., 2010). Running
promotional contests among fans who have to mention or tag their friends and
use a significant hashtag has a positive impact on the number of interactions and
network density (Naraine et al., 2019). Thus, sports organizations need to con-
tinue working on digital sport marketing with the purpose of increasing women’s
participation, and they can look for other ways of engagement outside the digital
platforms (Abeza et al., 2019). Filo et al. (2015) stated that the motives for the use
of social media by users may differ according to their socio-demographic charac-
teristics such as gender, age, educational level, or the sports context.
Finally, for sports events, social media communication channels allow the mes-
sage of sponsors to be amplified at sports events due to the network effect of users
sharing content with their supporters (Hazari, 2018). Chanavat and Desbordes
(2014) found that social media can be powerful advertising channels during a
sporting event such as the Olympic Games. This is corroborated by Burton’s
(2019) study, which believes that sponsor engagement has an illustrative compet-
itive advantage for brands through official sponsorship. However, other brands’
ambush marketing campaigns also have an optimistic view of consumer senti-
ment, so that success will depend on the delivery and effectiveness of the specific
partnership and the added value for consumers. For example, Gillooly et al. (2017)
demonstrated that sponsor brands achieve success in using social media when
they engage in reward-based activation. Finally, Hazari (2018) considers the use of
sponsorship as an important marketing tool to influence constructs such as sports
sponsorship, consumer attitudes, and purchase behavior.
This systematic review has several limitations. Firstly, the existence of hetero-
geneous lines of research makes it impossible to analyze a line of research related
90 S. Angosto et al.

to digital sports marketing in depth. A second limitation may be selection bias

since the diversity of studies may mean that some inclusion criteria are not suffi-
ciently clear and objective. Another limitation is the restriction of access to the
full text of certain studies that cannot be considered in the evaluation. Although
attempts have been made to consider different databases to carry out the system-
atic review, there may be a research bias by not considering possible studies pub-
lished in journals not indexed in the databases analyzed. Language bias should
also be considered by including only studies in English, excluding studies in other
languages such as Spanish or Portuguese. Finally, the last limitation is publication
bias since only studies with positive and significant results are usually published,
not studies with negative or non-significant results. These types of studies, if pub-
lished, tend to take longer to be published or are not widely cited.
Future lines of research can be developed based on the results of this systematic
review. Environment 3.0 offers many possibilities to be discovered (Tejedor et al.,
2019) as new technologies and digitalization are continuously growing and new
innovations are appearing. Most studies have focused on sports organizations,
with sports events not receiving as much attention, especially in the current con-
text of the COVID-19 pandemic. The limitation or prohibition of spectators at
sporting events represents a great business opportunity for brands as all fans must
follow and watch these events online in streaming or deferred. This creates an
ideal space to advertise and be able to evaluate consumer behavior toward the
advertising that is inserted in the online platforms during the event.
Among the different social media platforms studied, Instagram has not received
any attention. It is a platform that is gaining more importance and follow-up due
to its multimedia approach, highlighting the image over the text. The promotion
of products from different organizations through photos can be a study approach
that allows the number of interactions carried out on digital sport marketing to
be properly analyzed. The fitness sector is a very important area within current
sports management that has not received attention in the literature to evaluate
the possible digital marketing in this field. Other fields of study are digital sport
marketing in applications for the control and monitoring of physical activity or
management in sports centers. Finally, the existence of a bias according to gender
due to the lack of female supporters may represent an important opportunity for
future growth in the digital marketing of sports organizations (Naraine et al.,
2019).

References
Abeza, G., O’Reilly, N., & Seguin, B. (2019). Social media in relationship marketing:
The perspective of professional sport managers in the MLB, NBA, NFL, and NHL.
Communication & Sport, 7, 80–109.
Baena, V. (2018). The importance of CSR practices carried out by sport teams and its
influence on brand love: The Real Madrid Foundation. Social Responsibility Journal,
14(1), 61–79.
 Challenges in Sport Marketing 91

Burton, N. (2019). Exploring user sentiment towards sponsorship and ambush marketing.
International Journal of Sports Marketing and Sponsorship, 20(4), 583–602.
Chanavat, N., & Desbordes, M. (2014). Towards the regulation and restriction of ambush
marketing? The first truly social and digital mega sports event: Olympic Games, London
2012. International Journal of Sports Marketing & Sponsorship, 15(3), 2–11.
Dick, R. J., & Turner, B. A. (2007). Are fans and NBA marketing directors on the same
page? A comparison of value of marketing techniques. Sport Marketing Quarterly, 16(3),
140–146.
Evans, D. M., & Smith, A. (2004). Internet sports marketing and competitive advantage
for professional sports clubs: Bridging the gap between theory and practice. International
Journal of Sports Marketing and Sponsorship, 6(2), 9–21.
Filo, K., & Funk, D. C. (2005). Congruence between attractive product features and vir-
tual content delivery for Internet marketing communication. Sport Marketing Quarterly,
14(2), 112–122.
Filo, K., Funk, D. C., & Hornby, G. (2009). The role of web site content on motive and
attitude change for sport events. Journal of Sport Management, 23(1), 21–40.
Filo, K., Lock, D., & Karg, A. (2015). Sport and social media research: a review. Sport
Management Review, 18(2), 166–181.
García-Fernández, J., Elasri-Eijaberi, A., Pérez-Tur, F., Triadó-Ivern, X. M., Herrera-Torres,
L., & Aparicio-Chueca, P. (2017). Social networks in fitness centres: The impact of
fan engagement on annual turnover. Journal of Physical Education and Sport, 17(3),
1068–1077.
García Fernández, J., Fernández Gavira, J., Durán Muñoz, J., & Vélez Colón, L. (2015).
La actividad en las redes sociales: Un estudio de caso en la industria del fitness. Retos:
nuevas tendencias en Educación Física, deporte y recreación, 28, 44–49.
Gillooly, L., Anagnostopoulos, C., & Chadwick, S. (2017). Social media-based sponsorship
activation – A typology of content. Sport, Business and Management: An International
Journal, 7(3), 293–314.
Gómez, M., López, C., & Molina, A. (2019). An integrated model of social media brand
engagement. Computers in Human Behavior, 96(7), 196–206.
Hazari, S. (2018). Investigating social media consumption, sports enthusiasm, and gender
on sponsorship outcomes in the context of Rio Olympics. International Journal of Sports
Marketing and Sponsorship, 19(4), 396–414.
He, W., Zha, S., & Li, L. (2013). Social media competitive analysis and text mining: A
case study in the pizza industry. International Journal of Information Management, 3(3),
464–472.
Ioakimidis, M. (2010). Online marketing of professional sports clubs: Engaging fans on a
new playing field. International Journal of Sports Marketing & Sponsorship, 11(4), 271–282.
Johnston, M. A., & Spais, G. S. (2016). Sponsorship research: drawing on the past to
shape the future of sponsorship. In C. Campbell and J. J. Ma, (Eds), Looking forward,
looking back: drawing on the past to shape the future of marketing (pp. 734–735). New York,
NY: Springer.
Kane, G. C., Palmer, D., Phillips, A. N., Kiron, D., & Buckley, N. (2015). Strategy, not
technology, drives digital transformation. MIT Sloan Management Review and Deloitte
University Press, 14, 1–25.
Kannan, P. K., & Hongshuang, A. L. (2017). Digital marketing: A framework, review, and
research agenda. International Journal of Research in Marketing, 34(1), 22–45.
92 S. Angosto et al.

Kumar, L., & Bagchi, A. (2020). Comparative analysis of digital marketing strategies
for live streaming of sporting events. Annals of Tropical Medicine and Public Health, 23,
231–725.
López-Carril, S., Escamilla-Fajardo, P., González-Serrano, M. H., Ratten, V., & González-
García, R. J. (2020). The rise of social media in sport: A bibliometric analysis.
International Journal of Innovation and Technology Management, 17(06), 2050041.
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for
systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6(7),
e1000097.
Naraine, M. L., & Parent, M. M. (2017). This is how we do it: A qualitative approach to
national sport organizations’ social-media implementation. International Journal of Sport
Communication, 10, 196–217.
Naraine, M. L., Pegoraro, A., & Wear, H. (2019). #WeTheNorth: Examining an online
brand community through a professional sport organization’s hashtag marketing cam-
paign. Communication & Sport. https://doi.org/10.1177/2167479519878676
Phillips, E. (2015). Retailers scale up online sales distribution networks. The Wall
Street Journal November 17, 2015. Accessed from http://www.wsj.com/articles/
retailers-scale-up-online-sales-distribution-networks-1447792869
Popp, N., McEvoy, C., & Watanabe, N. (2017). Do college athletics marketers con-
vert social media growth into ticket sales? International Journal of Sports Marketing and
Sponsorship, 18(2), 212–227.
Schulz, K. F., Altman, D. G., & Moher, D. (2010). CONSORT 2010 statement: Updated
guidelines for reporting parallel group randomised trials. BMC Medicine, 340, c332.
Stavros, C., Meng, M. D., Westberg, K., & Farrelly, F. (2014). Understanding fan motiva-
tion for interacting on social media. Sport Management Review, 17, 455–469.
Tejedor, S., Cervi, L., & Gordon, G. (2019). Analysis of the structure and use of digital
resources on the websites of the main football clubs in Europe. Future Internet, 11(5),
104.
Tsitskari, E., Costa, G., Efthimis, K., Tzetzis, G., Quick, S., & Kioumourtzoglou, E. (2010).
Factors for evaluating the web as a marketing tool in Greek sport. Serbian Journal of
Sports Sciences, 4(3), 107–117.
Yoshida, M., Gordon, B., Nakazawa, M., & Biscaia, R. (2014). Conceptualization and
measurement of fan engagement: Empirical evidence from a professional sport context.
Journal of Sport Management, 28, 399–417.
Zanini, M. T., de Moraes, F. C., Lima, V., Migueles, C., Lourenco, C., & Irigaray, H. A. R.
(2019). Soccer and Twitter: Virtual brand community engagement practices. Marketing
Intelligence & Planning, 37(7), 791–805.
Chapter 5

Digital Ticketing Utilization

in High School Athletics
The Role of Spectator Demographics
Armin A. Marquez and Beth A. Cianfrone

Introduction
In the United States, high school football is a popular spectator sport. On any
given fall Friday night, when high school games are played, it is predicted that
seven million people are in attendance across the country (Niehoff, 2019). In
an environment where costs are continually rising, and revenue streams out-
side of school-assigned budgets are vital, high school athletic directors often rely
on ticket sales as a critical funding source (Hatfield & Hatfield, 2014). One strat-
egy that athletic directors have turned to increase ticket revenue is offering
digital ticketing. Digital ticketing refers to spectators’ online purchases (via phone
or website) and redemption for entry at the gate. Digital ticketing is common at
the professional level, and with the COVID-19 pandemic, it may become more
prevalent at other levels of sports, such as the high school level. While it is becom-
ing more common, many individuals prefer traditional walk-up ticket purchases
at the high school and/or many athletic directors who have not implemented
the digital ticketing option. There are many advantages to digital ticketing, so
understanding why some attendees do not utilize it is a worthwhile endeavor that
could assist administrators.
Sport administrators and marketers can use consumer data to target advertis-
ing or promotional materials to influence behavior. As part of this, they often
utilize market segmentation and start with understanding their consumers’
demographics. This information is often easily attainable by administrators
and can help inform decisions. Digital ticketing is a technology-based feature/
purchase, so specific demographic attributes may influence an individual’s
perception of the technology and intention to use the technology. As such, the
high school marketplace research is limited and warrants further discovery to
inform athletic directors of managerial decisions regarding digital ticketing (e.g.,
Cianfrone et al., 2015; Marquez et al., 2020a, 2020b). Even more specifically,
there is no research exploring the demographic influence on spectators’ adoption
of digital ticketing, which suggests the importance of the present study.
The purpose of this study is to explore differences among the factors influ-
encing the adoption of digital ticketing when attending high school football

DOI: 10.4324/9781003302476-5
94 A. A. Marquez and B. A. Cianfrone

games based on spectators’ demographics. Following the Theoretical Model for

Athletic Event Digital Ticketing Use (Marquez et al., 2020b), we assessed dif-
ferences in perceived usefulness, perceived ease of use, trust of digital ticketing,
willingness to pay convenience fees, and intention to use digital ticketing for
high school football games based on spectators’ demographic attributes (i.e.,
gender, age, income, education, race, affiliation to the school, and composi-
tion of the party in attendance). This information should be useful for athletic
administrators as they consider the adoption of digital ticketing and develop
marketing and ticketing strategies.

Theoretical Framework
Historically, high school athletic events, such as football games, require attend-
ees to purchase the game ticket at the stadium ticket booth with only cash
on the game day, minutes before start time, and then proceed to have their
ticket taken at the stadium entrance. Some challenges with this approach
include theft of the ticketing cash box due to insufficient security or use of vol-
unteer staff to manage the ticketing, crowd control from the various lines, and
spectator dissatisfaction with needing cash on hand to purchase the ticket in
an increasingly digital society (Marquez et al., 2020b). Further, advance ticket
sales may not be an option, and fear of the game selling out and long lines
may deter fans. For spectators, digital ticketing may improve their experience
by providing added flexibility in their purchases, including advance purchases,
and eliminating the need to have cash or waiting in ticket lines at the event.
For administrators, digital ticketing may resolve challenges associated with
cash gates, help secure revenue through advanced ticket sales, and provide
schools with the opportunity to collect customer data, which may prove useful
when devising sponsorship packages or marketing campaigns (Marquez et al.,
2020b). However, many attendees continue to utilize the cash walk-up purchase
option rather than digital ticketing in those high schools where both options
are offered. Thus, administrators need to assess if there are possible differences
across consumer segments.
Digital ticketing adoption by high school athletic departments has gained
momentum. The National Federation of State High School Associations (NFHS)
officially partnered with a ticketing company, Huddle Inc., to provide its member
schools with the option of adopting the digital ticketing platform GOFAN. The
company also works with schools on their traditional printed tickets (Huddle
Inc., 2017b). According to Griffin Pugh, General Sales Manager at Huddle, of
the nearly 9,000 schools using the Huddle’s physical printed tickets, about 11%
adopted the GOFAN platform for the 2018−19 school year (personal communi-
cation, August 30, 2018). Other digital ticketing companies are also in the high
school athletic space, such as Mascot Media, State Champs, Ticket Roar, and
Ticket Spicket, highlighting the attractiveness of this market (Marquez et al.,
2020b). While it is becoming more prevalent, some athletic directors do not
 Digital Ticketing Utilization 95

offer the option of digital ticketing to their fans and spectators who do not use
this technology, even when available. Marquez et al. (2020a) aimed to research
why athletic directors may or may not adopt digital ticketing for their programs.
Meanwhile, Marquez et al. (2020b) explored how attendees may perceive digital
ticketing over a series of technology attributes and their intentions to use digital
ticketing. Consideration of how digital ticketing factors are affected by potential
consumers’ demographic characteristics will add to the growing literature on dig-
ital ticketing.

Factors Influencing Adoption of Digital Ticketing

Researchers have used, modified, and extended the Technology Acceptance
Model (TAM; e.g., Davis, 1989, 1993) to examine the adoption of technology
by users in a variety of contexts. Many researchers have found its main factors
of perceived usefulness (capability to use the technology to one’s advantage)
and perceived ease of use (lack of effort to engage in the activity) influence
intention to use the technology (e.g., Davis, 1989, 1993; Venkatesh & Davis,
2000; Venkatesh et al., 2012). Fishbein and Ajzen (1975) explain that behavioral
intentions are a strong predictor of actual behavior. Most applicably, Marquez
et al. (2020b) adapted the TAM to explore spectators’ intentions to use digital
ticketing services when attending sporting events to derive a theoretical model
of athletic event digital use – Theoretical Model for Athletic Event Digital
Ticketing Use. They tested the influence of the variables perceived usefulness,
perceived ease of use, and intentions to use from the simplified TAM, and the
two new variables: trust of digital ticketing and the willingness to pay conven-
ience fees. An individual’s trust in digital ticketing referred to their trust in
the online environment to transmit sensitive information, such as credit card
numbers, to complete the ticket purchase. Willingness to pay considered spec-
tators’ willingness to pay extra (i.e., surcharges) to purchase digital tickets. They
found that trust, willingness to pay and perceived ease of use influenced specta-
tors’ purchase intentions, which, in turn, affected individuals’ intentions to use,
while willingness to pay also had a direct influence on intentions to use. While
the study was the first in the high school marketplace, it failed to address possible
demographical differences in individuals, which may be influencing perceptions
and intentions to use digital ticketing and could provide more information for
athletic directors. Because Marquez et al. (2020b) found that all factors influ-
enced intentions to use in some capacity, we include the five in our study in an
effort to segment the market.

Market Segmentation and Demographic Influences

Athletic directors may be concerned with understanding which spectators
are more inclined to use digital ticketing. Trail (2019) identifies demograph-
ics as one of the four primary bases for market segmentation, along with
96 A. A. Marquez and B. A. Cianfrone

psychographic, geographic, and behavioral factors. They also highlight that

using these bases in combination can result in a more comprehensive profile of
the consumer groups. In the technology space, Ha et al. (2007) and Venkatesh
and Davis (2000) have pointed to the importance of investigating the moderat-
ing effects of demographic factors. As such, we focus on demographic influences.
Various individual demographics could play a role in one’s adoption of tech-
nology, such as digital ticketing. Among the segments traditionally explored
in sports consumption or technology acceptance includes gender, age, income,
education, race, affiliation with the school, and the attendees’ party compo-
sition (with others, alone, etc.). For example, in 2008, Scarborough Research
identified Digital Savvy consumers as younger males with high education and
income levels (Scarborough Research, 2008). Yet, as digital technology becomes
more prevalent, this profile changes over the years and thus should be examined
in the sport digital ticketing context. Decision-makers may desire to know if
there are differences in these demographic groups on the five ticketing areas
and intentions to use.

Gender
Gender is often used as a point for segmentation in sport marketing literature,
with varying results. There are frequently behavioral differences between men
and women, such as in attendance or other forms of consumption. Yet, “research
does not support the existence of large differences between males and females
on things such as attitudes, beliefs, intentions, motives” (Trail, 2019, p. 289).
Therefore, perceptions on digital ticketing and intentions to use digital ticketing
may or may not differ by gender and needs to be investigated.

Age
One may expect differences in perceptions toward a technology based on age. Age
can influence the amount and type of media and technology people use. Millennials
have different technology habits than Baby Boomers (60+), Generations X
(45–55), and Y (26–44). As such, younger patrons may be more comfortable inter-
acting with technology, which would likely influence perceptions of ease, useful-
ness, trust of digital ticketing, acceptance of paying extra fees, and intentions to
use. Breaking down spectators’ perceptions toward and intention to use digital
ticketing by age may help identify those more likely to use the technology, provid-
ing greater insight than if we analyzed these elements separately.

Education
The level of education of an individual may also influence their perception
toward the technology. It is a common factor in determining attitudes and
behaviors (Hawkins & Mothersbaugh, 2010). Individuals pursuing higher
 Digital Ticketing Utilization 97

education levels may be more likely to gain exposure to technological advances,

which in turn could impact their perceptions, and perhaps be more likely to
leverage technology to achieve efficiencies in their everyday lives.

Income
Hawkins and Mothersbaugh (2010) noted that occupation and education
influence consumer preferences and purchase, whereas personal or household
income allows them to make the purchase. Although household income does
not lead to or explain consumer purchases, it is clear that it plays a role and
should be considered. Household income has been shown to explain some of
the variance in why certain consumers are willing to pay more for tickets (Popp
et al., 2018). Because there is an additional fee associated with digital ticketing,
perhaps income influences. Also, those with higher income are more likely to
engage with technology, influencing perceptions of ease, usefulness, and trust.
Those spectators reporting higher income levels may be more likely to see
traditional ticketing (e.g., long lines at a ticket gate) as a waste of their valu-
able time. Also, higher income may influence their willingness to pay fees for
the added convenience, directly impacting their intentions to use.

Race
It is unclear how or if racial identity influences individuals’ perceptions of digi-
tal ticketing technology. Trail (2019) notes that racial identity is rarely used in
research to distinguish differences within sport consumer behavior literature.
Nevertheless, finding such differences may provide valuable insights given that
very little is known about the high school athletic spectator profile.

Affiliation
High school sports may differ from professional sports in that attendees may
have unique ties to the school or athletic program. For example, Cianfrone
et al. (2015) found that most spectators in attendance of a high school bas-
ketball tournament were parents of the basketball players, but there are also
likely other spectator affiliations of attendees. School teachers or administra-
tors who have a direct affiliation to the school may be in attendance. Relatives,
beyond parents, of the participants (e.g., band members, cheerleaders, coaches,
athletes) and, finally, alumni of the school may also be in attendance. These
different affiliations may influence perceptions and intentions to use the dig-
ital ticketing option given their particular motives for attendance. Likewise,
spectators’ affiliation will likely affect the number of games attended during
a given season and their individual ticketing needs. Therefore, understand-
ing whether attendees’ adoption of the technology is linked to their affilia-
tion with the high school would enable athletic directors to design specific
98 A. A. Marquez and B. A. Cianfrone

marketing campaigns that align with each segment (e.g., “buy tickets for the
whole family in seconds with online ticketing”).

Party Composition
Who spectators choose to attend high school athletic events may influence
their purchasing needs and, therefore, their perceptions toward digital ticketing
and their intention to use this technology in the future. Perhaps groups are
attracted to the convenience of making one purchase and sharing tickets via
text messaging across group members (e.g., mom making the purchase and dis-
tributing tickets to children, grandparents, and other family members). On the
other hand, large family groups may wish to avoid paying convenience fees on
multiple tickets, which could add considerably to the expenses.
Identifying the different patterns of attendees and their ticketing preferences
can aid athletic directors and ticketing and marketing companies working in
the secondary education space to customize their efforts to educate fans about
the option of digital ticketing when purchasing tickets in advance for sporting
events. Additionally, from a practitioners’ standpoint, understanding the effect
of spectators’ demographic profiles on their perceptions toward digital ticketing
and future usage of this technology may help administrators design appropri-
ate marketing efforts. Therefore, it was considered applicable to measure the
differences in perceived usefulness, perceived ease of use, trust in digital tick-
eting, willingness to pay, and intentions to use based on spectator demograph-
ics (i.e., gender, age, income, education, race, affiliation, and attendee party
composition).

Method
As part of a larger data collection, after university IRB approval, we ascer-
tained data on adult spectators at 12 high school football games via a paper-
and-pencil survey using (1) a mall-intercept approach at various sections of the
stadium and (2) an information table, strategically located by concessions, at
five different stadiums, located in multiple types of communities (metropoli-
tan, suburban, and rural) across a southeastern state.
Participants completed 15 items on their digital ticketing perceptions and
intentions to use (i.e., perceived usefulness, perceived ease of use, trust in digi-
tal ticketing, willingness to pay fees, and intentions to use). The digital ticket-
ing factors were measured using multi-item 7-point Likert-type scale supported
by the literature (i.e., Davis, 1993; İkram & Cem, 2016; Taylor & Todd, 1995;
Zhao et al., 2016). Participants completed 7 demographic items (i.e., gender, age,
income, education, race, affiliation with the school, and who they attended with).
Age was divided into the categorical groupings used by Google Analytics (18–24,
25–34, 35–44, 45–54, 55–64, and 65-plus), as that information is accessible to
athletic directors. Participants identified their affiliation from a parent of high
 Digital Ticketing Utilization 99

school student, parent of event participants (i.e., coach, player, cheerleader, or

band member), non-parent family member (e.g., grandparents) of event partici-
pants, alumni of the high school playing in the game, high school representative
or staff member, or non-affiliated spectators (who noted their different affiliations,
such as football fan, community supporter).
To determine the overall differences in mean likelihoods between each of
the levels of the spectators’ demographic characteristics (i.e., independent var-
iables: gender, age, income, education, race, affiliation, and the composition of
the party in attendance), we performed seven multivariate analysis of variance
(MANOVA) on differences in ticket adoption factors (i.e., dependent variables:
perceived ease of use, perceived usefulness, trust of digital ticketing, willing-
ness to pay, and intention to use. Further, the results from the between-subject
effects test to identify the differences in the mean scores of the five dependent
variables for each independent variable, and a post hoc Tukey test to assess
pairwise comparisons between levels of those variables with more than two
groupings (age, income, education, race, affiliation, and party in attendance),
while applying a Bonferroni adjustment to account for the use of dependent
variables on multiple tests.

Results
On average, participants (N = 523) were 45 years of age, primarily White
(68.6%), married (75.5%), and earned a yearly family income of $100,000 or
higher (50.3%). The majority (>75%) had attended college, with 54.1% complet-
ing a four-year degree or higher. Half (49.9%) of the spectators reported being in
attendance with both adults and children, while the remainder were attending
with other adults (45.3%) or alone (4.8%). In terms of affiliation to the high
schools participating, 28.3% of respondents were alumni, 19.3% were non-parent
family members of event participants (i.e., players, cheerleaders, band members),
16.8% were parents of event participants, 14.7% were parents of high school
students (non-participants), 3.4% were representatives or staff of the school, and
17.4% did not identify with any of these categories. Participants reported having
attended an average of 6.76 (SD = 5.73) high school football games the previous
season. The items for the five digital ticketing factors showed good psychomet-
ric properties (α > .70). Participants reported highest on perceived usefulness
(M = 5.82, SD = 1.15), followed by perceived ease of use (M = 5.52, SD = 1.24),
trust in digital ticketing (M = 5.39, SD = 1.36), intention to use (M = 5.21,
SD = 1.52), and willingness to pay fees (M = 3.84, SD = 1.66).
To explore differences in ticketing adoption factors based on demographics
of the spectators, we used MANOVA, finding statistically significant differ-
ences based on gender [F(5, 517) = 2.76, p = .018], age [F(5, 517) = 4.60, p < .001],
school affiliation [F(5, 517) = 3.66, p = .003], and party attending [F(5, 517) = 2.26,
p = .047]. On average, female participants scored significantly higher on perceived
ease of use [F(1, 521) = 6.05, p = .014] and trust in digital ticketing [F(1, 521) = 4.26,
100 A. A. Marquez and B. A. Cianfrone

Table 5.1 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust

of Digital Ticketing, Willingness to Pay Convenience Fees, and
Intentions to Use by Gender

Construct Gender Participants M SD

Perceived ease of use Female 259 5.65* 1.29

Male 264 5.39* 1.19
Perceived usefulness Female 259 5.88 1.18
Male 264 5.76 1.11
Trust of digital ticket Female 259 5.52* 1.39
Male 264 5.27* 1.31
Willingness to pay convenience fee Female 259 3.98 1.81
Male 264 3.70 1.49
Intent to use digital ticketing Female 259 5.25 1.64
Male 264 5.17 1.41

Note: * Group significantly different (p < .05) from the others.

p = .039] than men (see Table 5.1). The post hoc Tukey tests assessed differences
on the factors of digital ticketing adoption based on independent variables with
three or more levels. Scheffe post hoc was used in cases where group sizes were
statistically different. Participants’ willingness to pay was found to be statistically
different based on age [F(5, 517) = 3.30, p = .006]. On average, participants of ages
ranging between 25–34 years scored significantly (p < .05) lower (M = 3.09, SD =
1.79) than the 55-64 (M = 4.36, SD = 1.76) and 65+ (M = 4.37, SD = 1.37) groups
(see Table 5.2).
Participants’ intention to use digital ticketing was found to be statistically dif-
ferent [F(4, 518) = 2.71, p = .029] based on their affiliation with the schools (see
Table 5.3). More specifically, the high school parents scored higher on intentions
to use (M = 5.60, SD = 1.17) than all other groups, although the difference was
only statistically significant compared to the alumni (M = 4.96, SD = 1.79) group.
Based on party attending, there were statistically significant differences on per-
ceived ease of use [F(2, 520) = 4.56, p = .01] and perceived usefulness [F(2, 520) =
5.17, p = .006]. Specifically, the parties comprised of adults and children (M = 5.64,
SD = 1.19; M = 5.96, SD = 1.03 respectively) scored higher than parties of just
adults (M = 5.34, SD = 1.29; M = 5.65, SD = 1.25 respectively), on both constructs
(Table 5.4). Meanwhile, differences among factors of digital ticketing perception
and adoption, based on ethnicity (Table 5.5), levels of income (Table 5.6), and
education (Table 5.7), were non-significant.

Discussion
Ticket sales are an essential revenue stream for many organizations, including
high school athletic departments. Digital ticketing offers schools an opportu-
nity to maximize advanced ticket sales efficiently and eliminate many potential
service issues (e.g., theft, lines, crowd control, etc.). To improve sport spectators’
 Digital Ticketing Utilization 101

Table 5.2 S pectators’ Perceived Ease of Use, Perceived Usefulness,

Trust of Digital Ticketing, Willingness to Pay Convenience
Fees, and Intentions to Use by Age Group

Construct Age Grouping Participants M SD

Perceived ease of use 18–24 27 5.52 0.93

25–34 38 5.54 1.29
35–44 113 5.68 1.15
45–54 272 5.48 1.26
55–64 43 5.32 1.31
≥65 30 5.46 1.52
Perceived usefulness 18–24 27 5.77 1.02
25–34 38 5.76 1.32
35–44 113 5.96 1.03
45–54 272 5.79 1.14
55–64 43 5.78 1.22
≥65 30 5.76 1.45
Trust of digital ticketing 18–24 27 5.36 1.33
25–34 38 5.26 1.41
35–44 113 5.51 1.25
45–54 272 5.38 1.34
55–64 43 5.35 1.58
≥65 30 5.38 1.61
Willingness to pay 18–24 27 3.59 1.73
25–34 38 3.09* 1.79
35–44 113 3.92 1.68
45–54 272 3.80 1.60
55–64 43 4.36* 1.76
≥65 30 4.37* 1.37
Intention to use 18–24 27 5.36 1.77
25–34 38 4.82 1.71
35–44 113 5.27 1.41
45–54 272 5.26 1.50
55–64 43 5.13 1.55
≥65 30 5.06 1.71

Note: * Group significantly different (p < .05) from the others.

experience, sport organizations must first understand their consumers and the
factors influencing their decisions (Trail, 2019). However, the lack of consumer
behavior research associated with interscholastic sports has left athletic direc-
tors, as well managers from marketing and ticketing companies working in
this space, to rely on trial and error or anecdotal information when making
critical decisions. The present study provides valuable insights into a market
segment that gets limited attention from researchers by exploring the role of
demographics.
Both men and women reported high intentions to use digital ticketing,
although there were no significant differences between the two. There were no
statistically significant differences found in participants’ perceived usefulness of
102 A. A. Marquez and B. A. Cianfrone

Table 5.3 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust of Digital
Ticketing, Willingness to Pay Convenience Fees, and Intentions to Use by
Affiliation

Perceived Ease of Use Participants M SD

High school parent 77 5.68 1.03

Parent of event participant 88 5.63 1.11
Family of event participant 101 5.52 1.26
Alumni of the high school playing in games 148 5.40 1.43
surveyed
High school representative or staff 18 5.63 1.05
Non-affiliated – Do not identify with a 91 5.43 1.21
specific HS
Perceived Usefulness Participants M SD
High school parent 77 6.05 0.88
Parent of event participant 88 5.84 1.06
Family of event participant 101 5.96 1.08
Alumni of the high school playing in games 148 5.65 1.34
surveyed
High school representative or staff 18 5.83 0.92
Non-affiliated – Do not identify with a 91 5.73 1.18
specific HS
Trust of Digital Ticketing Participants M SD
High school parent 77 5.55 1.12
Parent of event participant 88 5.38 1.35
Family of event participant 101 5.38 1.33
Alumni of the high school playing in games 148 5.30 1.60
surveyed
High school representative or staff 18 5.74 1.00
Non-affiliated – Do not identify with a 91 5.37 1.22
specific HS
Willingness to Pay Participants M SD
High school parent 77 3.99 1.53
Parent of event participant 88 3.66 1.59
Family of event participant 101 3.92 1.68
Alumni of the high school playing in games 148 3.82 1.75
surveyed
High school representative or staff 18 3.57 1.49
Non-affiliated – Do not identify with a 91 3.88 1.70
specific HS
Intention to Use Participants M SD
High school parent 77 5.60* 1.17
Parent of event participant 88 5.22 1.38
Family of event participant 101 5.39 1.39
Alumni of the high school playing in games 148 4.95* 1.79
surveyed
High school representative or staff 18 5.22 1.37
Non-affiliated – Do not identify with a specific HS 91 5.11 1.57

Note: * Group significantly different (p < .05) from others.

 Digital Ticketing Utilization 103

Table 5.4 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust

of Digital Ticketing, Willingness to Pay Convenience Fees, and
Intentions to Use by Party Attending

Construct Party Attending Participants M SD

Perceived ease of use Alone 25 5.85 1.10

Adults 237 5.34* 1.29
Adults and children 261 5.64* 1.19
Perceived usefulness Alone 25 6.03 1.12
Adults 237 5.65* 1.25
Adults and children 261 5.96* 1.03
Trust of digital ticketing Alone 25 5.51 1.37
Adults 237 5.24 1.45
Adults and children 261 5.52 1.26
Willingness to pay Alone 25 3.96 1.97
Adults 237 3.75 1.61
Adults and children 261 3.91 1.68
Intentions to use Alone 25 5.57 1.38
Adults 237 5.04 1.61
Adults and children 261 5.33 1.44

Note: * Group significantly different (p < .05) from others.

digital ticketing or willingness to pay fees based on gender. However, women

scored statistically significantly higher than men on perceived ease of use and
trust of digital ticketing. Although the difference was statistically significant, the
men still reported strongly (>5.0) on both factors, suggesting that administrators
should not segment their marketing materials about digital ticketing by gender.
If women have more experience making purchases, they may find it easier and
more useful to leverage technology to achieve efficiencies. Experience may also
influence trust toward the use of technology, ultimately determining intention to
use digital ticketing in the future.
In examining the role of age on the various ticketing factors, we found dif-
ferences in participants’ willingness to pay fees associated with digital ticketing.
Those over 55 years old (55–65 and 65+) were more likely to be willing to pay the
surcharges associated with digital ticketing than the other age groups. Individuals
in the remaining age groups reported means below the 4.0 midpoint, indicat-
ing they were unwilling to pay the fees. Perhaps, older participants value their
time more and would rather pay extra than stand in long lines. This group may
also have higher disposable income. Although no differences were found based
on income levels, they may have higher discretionary income to utilize on ser-
vices like digital ticketing. Targeted marketing materials could be geared toward
this demographic. More importantly, it is clear that fees affect consumers’
perceptions toward digital ticketing, so searching for alternatives to remove this
obstacle is critical. For example, some schools have searched for ways to absorb
the fees, such as adjusting prices at the gate to match the total cost of digital
purchases. However, there is no empirical evidence to show whether this strategy
has been successful in improving patrons’ perceptions toward the technology.
104 A. A. Marquez and B. A. Cianfrone

Table 5.5 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust of Digital
Ticketing, Willingness to Pay Convenience Fees, and Intentions to Use by
Race

Construct Age Grouping Participants M SD

Perceived ease of use White 359 5.51 1.22

Black 96 5.63 1.16
American Indian/Alaska Native 4 5.50 1.55
Hawaiian/Pacific Islander 3 6.56 0.51
Some other race 10 5.27 1.61
Not answered 51 5.32 1.44
Perceived usefulness White 359 5.85 1.15
Black 96 5.83 1.09
American Indian/Alaska Native 4 5.83 1.45
Hawaiian/Pacific Islander 3 6.56 0.51
Some other race 10 5.07 1.65
Not answered 51 5.73 1.16
Trust of digital ticketing White 359 5.44 1.33
Black 96 5.52 1.23
American Indian/Alaska Native 4 5.58 1.64
Hawaiian/Pacific Islander 3 4.67 2.52
Some other race 10 4.63 1.65
Not answered 51 4.99 1.55
Willingness to pay White 359 3.82 1.75
Black 96 4.06 1.36
American Indian/Alaska Native 4 4.58 1.62
Hawaiian/Pacific Islander 3 2.89 1.02
Some other race 10 3.17 1.64
Not answered 51 3.69 1.53
Intentions to use White 359 5.22 1.55
Black 96 5.31 1.33
American Indian/Alaska Native 4 5.58 1.34
Hawaiian/Pacific Islander 3 6.44 0.96
Some other race 10 4.10 2.11
Not answered 51 5.07 1.52

In terms of affiliation to the school, high school parents were most likely
to use digital ticketing (highest intentions). This group may already perform
other school-related functions digitally (e.g., registering their children to
participate in specific activities, paying for school lunch, and even purchas-
ing school supplies). This insight would seem to have substantial marketing
strategy repercussions, but fortunately for athletic administrators, all groups
reported positively intending to use digital ticketing. Interestingly, alumni of
the school ranked lowest on intention to use digital ticketing. Perhaps for this
group, they are not aware of the option because it was not available when they
were in school.
Some digital ticketing preferences differed by the groups who attended the
games. The attendance parties composed of both adults and children (e.g., parents
 Digital Ticketing Utilization 105

Table 5.6 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust

of Digital Ticketing, Willingness to Pay Convenience Fees, and
Intentions to Use by Income

Construct Age Grouping Participants M SD

Perceived ease of use Less than $20,000 11 5.48 0.94

$20,000–$34,999 17 5.94 0.85
$35,000–$49,999 33 5.51 1.07
$50,000–$74,999 65 5.46 1.25
$75,000–$99,999 74 5.59 1.36
$100,000 or more 263 5.53 1.26
Not answered 60 5.32 1.25
Perceived usefulness Less than $20,000 11 5.94 0.89
$20,000–$34,999 17 6.18 0.73
$35,000–$49,999 33 5.94 1.09
$50,000–$74,999 65 5.59 1.21
$75,000–$99,999 74 5.83 1.32
$100,000 or more 263 5.87 1.11
Not answered 60 5.66 1.20
Trust of digital ticketing Less than $20,000 11 5.73 0.92
$20,000–$34,999 17 5.94 0.88
$35,000–$49,999 33 5.40 1.12
$50,000–$74,999 65 5.10 1.41
$75,000–$99,999 74 5.54 1.44
$100,000 or more 263 5.41 1.33
Not answered 60 5.22 1.56
Willingness to pay Less than $20,000 11 3.85 1.66
$20,000–$34,999 17 3.65 1.66
$35,000–$49,999 33 3.89 1.86
$50,000–$74,999 65 3.85 1.68
$75,000–$99,999 74 3.82 1.69
$100,000 or more 263 3.85 1.63
Not answered 60 3.82 1.68
Intentions to use Less than $20,000 11 5.91 0.88
$20,000–$34,999 17 5.45 1.22
$35,000–$49,999 33 5.45 1.29
$50,000–$74,999 65 4.95 1.67
$75,000–$99,999 74 5.18 1.71
$100,000 or more 263 5.25 1.45
Not answered 60 5.04 1.69

attending with high school students or younger children) had higher perceived
ease of use and perceived usefulness of digital ticketing than parties of just adults.
Attending sporting events with children can pose different challenges than when
attending with only adults. Perhaps those attending with children are accustomed
to using technology to their advantage when faced with inconvenient tasks, like
standing in line to purchase tickets at a box office and remembering to have
cash on hand and therefore find it both more useful and easier to use. As such,
athletic directors or ticketing companies could market with images of those with
106 A. A. Marquez and B. A. Cianfrone

Table 5.7 S pectators’ Perceived Ease of Use, Perceived Usefulness, Trust

of Digital Ticketing, Willingness to Pay Convenience Fees, and
Intentions to Use by Education

Construct Age Grouping Participants M SD

Perceived ease of use High school 52 5.47 1.24

College 316 5.53 1.22
Master’s 70 5.70 1.29
Post-graduate 18 5.80 1.17
Not answered 67 5.21 1.28
Perceived usefulness High school 52 5.80 1.04
College 316 5.88 1.09
Master’s 70 5.78 1.36
Post-graduate 18 6.07 0.86
Not answered 67 5.52 1.30
Trust of digital ticketing High school 52 5.40 1.38
College 316 5.42 1.34
Master’s 70 5.36 1.45
Post-graduate 18 5.41 1.43
Not answered 67 5.27 1.36
Willingness to pay High school 52 4.20 1.27
College 316 3.80 1.70
Master’s 70 3.57 1.78
Post-graduate 18 3.83 1.76
Not answered 67 4.03 1.54
Intentions to use High school 52 5.25 1.43
College 316 5.24 1.51
Master’s 70 5.14 1.80
Post-graduate 18 5.33 1.38
Not answered 67 5.10 1.40

families at games and relay the usefulness and ease of use of digital ticketing to
those groups.
The participants did not differ in their perceptions or intentions to use dig-
ital ticketing based on their ethnicity, level of income, and education. It was
somewhat surprising that levels of income did not play a role, given the rela-
tionship between technology access and wages and previous literature about
the role income may play in consumption (e.g., Hawkins et al. 2010). Similarly,
an individual’s education level sometimes predicts perceptions or intentions to
use, so it was interesting to see that lack of influence. However, digital ticketing
prevalence in other aspects of lives (e.g., movies, professional sports, plays, con-
certs) may span income and education levels. Non-statistical differences based
on ethnicity are still valuable, conveying less need for differentiation in market-
ing messaging.
Although we identified statistical differences based on spectators’ character-
istics, the magnitude of those differences was not distinctive enough to extrap-
olate practical conclusions that would allow schools and marketing companies
to differentiate their messaging depending on the targeted recipients. Therefore,
 Digital Ticketing Utilization 107

based on this study, digital ticketing is the type of service that does not need to
be targeted at a specific segment. As such, athletic directors should direct their
marketing efforts broadly and not try to segment or target certain genders, ages,
or types of attendees based on their affiliation with the school or with whom they
attend games. Future researchers should explore the other bases for market seg-
mentation (i.e., psychographic, geographic, and behavioral factors; Trail, 2019).
Nevertheless, regardless of the individual characteristics, spectators reported high
intentions to use digital ticketing when attending high school athletic events if
given the opportunity. Such a finding can provide athletic directors with a cer-
tain degree of confidence in their decision to adopt the technology.
It is worth noting that the sample represents adult high school football spec-
tators, and despite the popularity of high school football, spectators across other
interscholastic sports may differ in perceptions toward and intentions to use dig-
ital ticketing. Also, some study participants had experienced digital ticketing in
their respective high schools when attending sporting events, while others did
not. Such experiences may have spurred individual differences in perceptions
toward the technology and high schools’ capabilities to execute its implementa-
tion. Future research may consider a more diverse sample across multiple states
while accounting for various sports. Another valuable area for exploration may
be high school students and their ticket purchasing behaviors. Similarly, factors
influencing the decision to adopt digital ticketing in high school athletic depart-
ments warrant further exploration. Perhaps considering the relationship between
adoption of multiple technological advances within high school athletic depart-
ments, such as video analysis for team analysis and scouting opposition, and the
likelihood of digital ticketing adoption.

References
Cianfrone, B. A., Zhang, J. J., Pitts, B., & Byon, K. K. (2015). Identifying key market
demand factors associated with high school basketball tournaments. Sport Marketing
Quarterly, 24, 91–104.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of
information technology. Management Information Systems Quarterly, 13, 319–336.
Davis, F. D. (1993). User acceptance of information technology: System characteristics,
user perceptions, and behavioral impacts. International Journal of Man-Machine Studies.
38, 475–487.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to
theory and research. Reading, MA: Addison-Wesley Pub. Co.
Ha, I., Yoon, Y., & Choi, M. (2007). Determinants of adoption of mobile games under
mobile broadband wireless access environment. Information & Management, 44(3),
276–286.
Hatfield, L. H., & Hatfield, L. C. (2014). Understanding corporate sponsorships: Moving
from a $100 ad purchase to a $1,000 sponsorship agreement. Strategies, 27, 37–41.
Hawkins, D. I., & Mothersbaugh, D. L. (2007). Consumer behavior: Building marketing
strategy (11th ed.). Boston, MA: Irwin/McGraw-Hill.
108 A. A. Marquez and B. A. Cianfrone

Huddle Inc. (2017a). Big news for Rhode Island high schools and sport fans. Retrieved
from: http://www.huddleinc.com/blog/rhode-island-interscholastic-league-selects-
huddle-inc-for-exclusive-digital-ticketing-partnership
Huddle Inc. (2017b). Huddle, the nation’s leading high school event ticketing provider,
secures digital event ticketing rights to the Western Pennsylvania Interscholastic
Athletic League (WPIAL). Retrieved from: https://www.prnewswire.com/news-releases/
huddle-the-nations-leading-high-school-event-ticketing-provider-secures-digital-
event-ticketing-rights-to-the-western-pennsylvania-interscholastic-athletic-
league-wpial-300467750.html
İkram, D., & Cem, G. (2016). Factors affecting the adoption of mobile payment systems:
An empirical analysis. Emerging Markets Journal, 6, 16–24.
Kim, Y., Kim, S., & Rogol, E. (2017). The effects of consumer innovativeness on sport
team applications acceptance and usage. Journal of Sport Management, 31(3), 241–255.
Marquez, A. A., Cianfrone, B. A., & Kellison, T. (2020a). Factors affecting leaders’ adop-
tion of innovation: The case of digital ticketing in the high school athletic space. Sports
Innovation Journal. http://journals.iupui.edu/index.php/sij
Marquez, A. A., Cianfrone, B. A., & Kellison, T. (2020b). Factors affecting spectators’
adoption of digital ticketing: The case of interscholastic sports. International Journal of
Sport Marketing and Sponsorship, 21(3), 527–541.
National Federation of State High School Associations (NFHS). (2019). 2018-19 High
school athletic participation survey. Retrieved from: https://www.nfhighschool.org/
media/1020412/2018-19_participation_survey.pdf
Niehoff, K. (2019, September 25). The NFHS voice: High school football is thriving, not
dying. National Federation of State High School Associations. Retrieved from: https://www.
nfhighschool.org/articles/the-nfhigh school-voice-high-school-football-is-thriving-
not-dying/
Popp, N., Shapiro, S., Walsh, P., McEvoy, C., Simmons, J., & Howell, S. (2018). Factors
impacting ticket price paid by consumers on the secondary market for a major sporting
event. Journal of Applied Sport Management, 10(1), 23–33.
Scarborough Research (2008). Understanding the digital savvy consumer. www.scarborough.
com
Taylor, S., & Todd, P. A. (1995). Understanding information technology usage: A test of
competing models. Information Systems Research, 6(2), 144–176.
Trail, G. T. (2019). Sport consumer behavior (4th ed.). Seattle, WA: Sport Consumer
Research Consultants LLC.
Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the Technology
Acceptance Model: Four longitudinal field studies. Management Science, 46(2), 186–204.
Venkatesh, V., Thong, J. Y., & Xu, X. (2012). Consumer acceptance and use of informa-
tion technology: Extending the unified theory of acceptance and use of technology. MIS
Quarterly, 36(1), 157–178.
Zhao, Q., Chen, C., Wang, J., & Wang, K. (2016). Study of factors influencing mobile
ticketing adoption: Status quo bias perspective. Journal of Marine Science and Technology,
24, 926–937.
Chapter 6

Social Media and Stakeholder

Relationships
Analyzing an Online Communication
Network in the Sport Industry
Juha Yoon, Wonseok (Eric) Jang, and Paul M. Pedersen

Introduction
Sports in general are popular in South Korea (hereafter Korea as shorthand). The
revenue generated from the sports and the outdoor activity market in Korea is
$3,095 million in 2020 and is expected to increase $4,131 million by 2025 (Statista,
n.d.). In addition, Korea has hosted several international sporting events, such
as the 1988 Summer Olympics, 2018 Winter Olympics, and 2002 World Cup.
Moreover, the country is also home to several professional sports leagues, which
further suggest the popularity of sport. In Korea, the most popular professional
sports league is a baseball league. The Korean Baseball Organization (KBO)
league, which was introduced in 1982, has enjoyed increased interest and fan sup-
port over the last decade. This is evidenced by the fact that the price for the rights
to broadcast KBO was just under $300,000 in 2010 and had increased to over
$33 million by 2015 (Lee, 2017). In addition, the total economic impact of the
KBO reached over $1 billion in 2011 (Kim, 2016). Furthermore, reports generated
by the KBO have indicated that the number of league spectators has also increased
significantly in the last decade. In 2007, the number of spectators was approxi-
mately 4.1 million, and that number increased to 8.4 million by 2017 (KBO, n.d.).
The second most popular professional sports league in Korea is a football
league (K League), which was introduced in 1983. While soccer was at one time a
sport that garnered interest in Korea, interest in and popularity of the K League
has declined over the last decade. In 2007, the total number of spectators from
14 teams was 2.7 million, and that number from 22 teams had decreased to
1.5 million by 2018 (K League Information System, n.d.). In addition, in 2015, the
price for the rights to broadcast K League was just over $4.1 million (Kim, 2015),
which was only 12.4% of the price for the right to broadcast KBO.

The Golf Industry in South Korea

In Korea, golf is considered a luxury sport due to the high cost of a round of
golf. Despite the relatively steep financial outlay, golf’s popularity has steadily
increased among Koreans since 1988. In 1988, Se-Ri Pak, a Korean golf legend,
made history when she won the LPGA (Ladies Professional Golf Association)
DOI: 10.4324/9781003302476-6
110 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

U.S. Women’s Open. Pak’s win served as the catalyst to the golf boom in Korea.
Many children who are called as “Se-Ri Kids” were inspired by Se-Ri Pak’s victory
and were thus encouraged to learn and play golf in hopes that they might become
the world’s next best golfer (Randall, 2012).
The Korea Golf Course Business Association (2017) announced that,
as of 2016, Korea was home to 486 golf courses (the total number of visitors =
36.7 million), 236 more golf courses than had existed in the country in 2006
(n = 250, the total number of visitors = 19.7 million). In addition, the Korean
Golf Association (KGA, 2017) found that approximately 6.36 million people
played golf at least once in 2017, which was a 253% increase over the number of
people who had played golf in 2007 (n = 2.51 million). Those golfers each spent
around $300 per month on golf-related purchases (e.g., green fees or golf equip-
ment). Moreover, Korean golfers even travel to overseas to play golf. According
to the KGA (2017), roughly 2.11 million people had traveled to foreign countries
(e.g., Thailand, Philippines, China) to play golf in 2017. In 2007, fewer than
600,000 Korean golfers (n = 560,000) took part in overseas golf tours.
The Korea Ladies Professional Golf Association (KLPGA) tour is one of the
most popular spectating sports in Korea (Yoo, 2012). The KLPGA was intro-
duced in 1978, and as of December 2020, there are 2,716 members (KLPGA,
n.d.-a). In 2019, the KLPGA hosted 30 KLPGA tournaments, which is an
increase of approximately 50% compared to the number of tournaments the
association hosted in 2009 (n = 20; KLPGA, n.d.-b). The success of women’s golf
in Korea is not limited to the KLPGA. Korean women golfers are also dominat-
ing the LPGA tour. Since Ok-Hi Ku claimed Korea’s first LPGA victory in 1988,
100 more Korean women golfers, as of 2011, went on to win the association’s
tournaments (Koo, Kim, & Won, 2014). The success of Korean women golfers is
even greater today. One of the “Se-Ri Kids”, Inbee Park became the first female
gold medalist in the history of Olympics at the inaugural women’s golf event,
the 2016 Rio Summer Olympics. In 2019, Korean women golfers won 15 of the
LPGA’s 33 tournaments (45%). Similarly, Korean women golfers won approxi-
mately 43% (15 of 35) of the LPGA tournaments in 2017 (LPGA, n.d.). Because
of this success, 11 Korean women golfers are ranked in the top 25 in the Rolex
World Ranking as of February 2019 (Rolex Ranking, n.d.).
Despite this success, the KLPGA tour suffered a dearth of title sponsors in
the early day (1980–1990s) (KLPGA, n.d.-c). Korea’s economy was exposed to
and was hurt by corruption in the 1990s, and because golf was considered a
luxury sport, there was a lack of interest in golf during that time. Even though
the KLPGA tour secured its first title sponsor in 1990, the association contin-
ued to suffer from a lack of financial support throughout the 1990s. Despite the
difficulty experienced in that early period, however, interest in women’s golf in
Korea has increased substantially, especially during the last decade. In 2017, the
KLPGA hosted 32 tournaments, and all of those tournaments were sponsored by
major organizations (KLPGA, n.d.-b). In addition, all tournaments were broad-
casted by Seoul Broadcasting System Golf (SBS Golf). Between 2014 and 2016,
 Social Media and Stakeholder Relationships 111

SBS paid roughly $4 million annually to broadcast the tournament (Han, 2013).
Furthermore, in 2017, approximately 278,000 golf fans visited the KLPGA tour-
nament to watch the professional golfers (Kwon, 2017). Because of the popularity
of women’s golf in Korea, several corporations (e.g., Hi-Mart, CJ Golf, Hana
Bank) have begun sponsoring a variety of golf professionals. Major corporations
are especially interested in sponsoring golf professionals because those corpora-
tions are able to maximize the effects of their sponsorships by investing only a
relatively small amount of money in the sponsored players (Moon, 2006).
Another unique characteristic of the Korean National Golf Tour is that
female golfers have received more interest from fans and sponsors than male
golfers (Kim & Seo, 2018). For example, in 2014, 101 out of 137 Korean LPGA
golfers (73.7%) have endorsement contracts with primary sponsors, while only
64 out of 144 Korean PGA golfers (44.4%) have similar contracts with pri-
mary sponsors (Han, 2014). In addition, the television rating is higher for the
KLPGA tour compared to the KPGA. This is unique in that many other coun-
tries (e.g., the USA, several European countries) tend to focus their attention
on male athletes and male-dominated sports.

Media Coverage of Golf in South Korea

Due to the symbiotic nature of sport and the media, mediated forms of com-
munication (e.g., mass media, social media) have assisted sport organizations
in generating interest among and rapport with sport fans across the globe.
Korea, as well as much of the rest of the world, is rich in sport-related news
that originates with a plethora of media outlets, both traditional and digital in
nature, including social media platforms (Yoo, 2012). Further, Korea has real-
ized a growth rate regarding social media adoption that is largely unmatched
elsewhere in the world (Lee et al., 2018). The Korean Internet and Security
Agency (KISA) conducted a survey that found that, in 2016, 65.2% of Korean
Internet users regularly used social networking sites. Further, the agency found
that over 90% of Korean Internet users accessed these sites using their smart-
phone devices (KISA, 2016). South Korea’s widespread adoption of social media
is said to be perhaps due to the country’s high-speed Internet, free Wi-Fi access,
and ubiquitous nature of smartphones (Choi & Park, 2014). Nevertheless, the
KLPGA’s social media use is significantly underused.
As a result of Korea’s economic growth, people’s desires to participate in sports
and leisure have increased. Accordingly, people’s interest in golf has increased.
Further, after Ok-Hi Ku achieved victory at the LPGA of Japan Tour (JLPGA)
in 1984, and after she passed the LPGA qualifying school in 1985, the media in
general (e.g., magazines, newspapers) became more focused on golf (Kim & Seo,
2018). As such, the development of golf in Korea is closely related to the media
attention it has garnered. That is, through the media, professional golf channels
became available to consumers, and consumer interest led to increased compe-
tition with regard to broadcasting rights.
112 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

Since 1998, there has been a gradual increase in the number of articles
published that have named KLPGA golfers. Lee and Jung (2016) found that
4,447 articles including the keywords or phrases “Korean LPGA golfer”, “Korean
female golfer”, and “LPGA” have been published in established Korean newspa-
pers (i.e., Chosun-Ilbo and Dona-Ilbo) beginning the year Pak made her debut to
August 2015. These articles have primarily focused on performance, achievement,
nationality, physical beauty, gender role, and femininity. This has resulted in some
prejudice and gender-based discrimination on the part of the media with regard to
KLPGA. Still, KLPGA athletes continue to use various media outlets, including
social networking platforms, to engage with their Korean and global fans.

Social Media Usage by the KLPGA and LPGA

To effectively reach out to existing and potential fans, KLPGA and LPGA
actively utilize social media, such as Facebook and Twitter. The development
of various social media platforms has profoundly changed how people consume
and discuss sport, as well as how stakeholders (e.g., fans, athletes, sponsors, event
organizing committees, teams) interact with one another. More specifically,
social media platforms can allow for, among other things, live media coverage
of sport events, streaming broadcasts of the events, fan polls, live interviews
with both athletes and commentators, and behind-the-scenes looks at things
like practices and pre-game rituals. Because social media platforms function in
a variety of ways, stakeholders are increasingly interested in employing them
as means of keeping up to date with their favorite athletes, teams, and events
(Pfanner, 2012). In this regard, previous studies have examined the effects of
different ways of using social media to communicate with their stakeholders (e.g.,
fans, sponsors) from individual athletes’ (e.g., Lebel & Danylchuk, 2012; Thorpe,
2017), sport national governing bodies’ (e.g., Eagleman, 2013; Naraine & Parent,
2016), and professional sport teams’ perspectives (e.g., Gibbs et al., 2014).
Social media has proven valuable to golf courses as well as other golf organ-
izations because it enables them to share content with customers and mem-
bers and to therefore promote or grow their online communities (Tracy, 2018).
For example, the LPGA makes use of social media platforms, and as of December
2017, more than 178,000 people were following the LPGA (@LPGA) on Twitter.
The LPGA’s senior director of social media marketing and communications,
Tina Barnes-Budd, emphasized the importance of using social media. In dis-
cussing the importance of social media to the LPGA, she discussed the ways
in which various social media platforms had become a means through which
the LPGA could promote both golfers and tournaments and noted that social
media had permitted the LPGA to better interact with fans (White, 2018). To
promote golfers and tournaments, the LPGA has taken to including photos
and videos in their social media posts because visuals such as these tend to be
popular among fans across the globe. The LPGA’s manager of digital content
at the LPGA, Jennifer Meyer, also emphasized the role social media has played
 Social Media and Stakeholder Relationships 113

in providing organizations such as the LPGA with opportunities to get their

messages out and to encourage growth among their impassioned and invested
fan bases. Social media also presents opportunities to fans as it enables them to
both connect with organizations such as the LGPA and to become players in
the organizations’ stories (Meyer, 2017).

Purpose of Study
Sport has become a true indicator of globalization. Throughout the world,
sport-related interactions have increased as a result of relatively recent advances
in technology. This study investigated whether extant trends shifted with regard
to the KLPGA as compared with the LPGA and each organization’s social media
communication networks. Upon examining the Twitter communication network,
it becomes apparent that not only trends but also consumer attitudes can suggest
what types of relationship might exist between sports in general and social media
outlets. This is largely due to the fact that social media both reflects social reality
and can establish and define social reality. Therefore, the purpose of this study
is to examine the online social networks affiliated with the KLPGA and LPGA
to investigate the online communication relationships that exist among users, as
this will help to enhance our understanding with regard to who or what serves as
influential stakeholders in the digital environment. The research question that
guided this study asked the following: What degree do KLPGA and LPGA stake-
holders engage with Twitter. In an attempt to answer this research question, this
study investigated the usage of Twitter by key LPGA and KLPGA stakeholders,
their networks, the groups involved, and the differences in position advantages of
the groups involved.

Method
The methodology for this study was developed based on social network con-
cepts. The Python programming language was employed per target user (i.e.,
@korealpga and @LPGA) in order to create Python codes, which were used to
extract tweets from two official Twitter pages (i.e., those of the KLPGA and
LPGA). In addition, as means of enhancing our understanding of the public’s
engagement level, data regarding the number of likes and retweets were collected.
Lastly, the structure and position of the users within the KLPGA and LPGA
networks were identified by extracting all of the usermentions in the tweets pub-
lished on the official KLPGA or LPGA Twitter accounts. Regarding the users
and ties in this investigation, mentioned users within tweets to the KLPGA
(@korealpga) or LPGA (@LPGA) Twitter accounts were the nodes, and the rela-
tionships connected to the usermentions were the ties. There was directional-
ity in the usermentions relationships as the arrowheads indicated the amount of
backward and forward flow that existed within each of the various pairs of nodes.
Additionally, the authors of this study sought to examine the network ties with
114 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

regard to their strengths and weaknesses. As such, data associated with the ties
were weighted based on mentioning frequency.
Adjacency matrices were used so that the networks under study could be
conceptualized mathematically (Borgatti et al., 2013). Each relationship that
exists between any given pair of actors in the network is mapped out via a
square matrix. With a large amount of data, correctly entering all the values
into the cells would have been practically impossible to do manually. Thus,
adjacency matrices were made through the creation of Python codes. As noted
by Borgatti and colleagues, social network analysis is used to examine not only
networks but also the attributes of the nodes, which are what differentiate
the nodes from one another. Some nodes will share similarities, however, and
these similarities can be defined in terms of nodes’ attributes. Via these simi-
larities or common attributes, one can identify both central groups and cohe-
sive subgroups; this makes it also possible to assess whether subgroups might
become central groups. Further, an attribute matrix and an adjacency matrix
were created. This enabled us to determine that a total of eleven heterogeneous
types of users exist. These include women golfers, men golfers, golf courses or
clubs, golf tournaments, sponsors or partnerships, other sport players or organ-
izations, golf associations, celebrities, media-related stakeholders, public users,
and non-sport-related organizations or sponsors. The nodes were then cate-
gorized based on the eleven types of users, as this allowed for identification
of the KLPGA and the LPGA’s primary online stakeholder groups and their
interactions.
In order to better understand the usermentions structure of interaction pat-
terns, a one-mode network was assessed. Usermentions relationships permitted
the authors of the study to examine how the actors were tied to one another.
The application of graph theory principles affiliated with nodes and edges
was done to build a usermentions interaction network. In order to answer the
research question, with imported (adjacency) and attributional (KLPGA and
LPGA) matrices, UCINET was used. Exploring quantitative measures (e.g.,
in-degree, out-degree, beta centrality) was facilitated through the use of this
social network software. This was predicated on the identified interactive rela-
tionships between the nodes and clusters formed by specific types of nodes,
resulting in sub-networks within the perceived domain. This study also made
use of social network visualization software (i.e., NetDraw), which assisted in
the interpretation of the data and results. Once the network was created, all
users were color-coded by network user type. Further, the interactivity strength
was highlighted by the degree to which the node size was set. As such, the
greater the degree centrality, the larger the node was scored. The networks
were then rearranged visually by organization type, as this enabled the iden-
tification of interactions (i.e., level) and stakeholders (i.e., primary) affiliated
with the organizations. All of this permitted the viewing of a more fully
assessed social media (i.e., Twitter and usermentions) network and structure of
communication.
 Social Media and Stakeholder Relationships 115

Results
To address the research question for this study, data published from August 2,
2010 to December 7, 2017 were collected from KLPGA and LPGA’s official
Twitter accounts. The KLPGA had posted 2,922 tweets on its Twitter web-
page (i.e., @korealpga) and the organization had 3,978 followers. In the same
time span, the LPGA (i.e., @LPGA), however, had posted 39,300 tweets and
had 178,000 followers. For this study, all 2,922 tweets published on the offi-
cial KLPGA Twitter webpage between August 2, 2010, when the account was
created, and December 7, 2017, were collected using Python programming lan-
guage. Due to the limitations associated with collecting data via Python, only
3,162 tweets published on the LPGA webpage were collected over the final five
months of the data collection period (Table 6.1). In addition to the differences
in the number of published tweets, there are significant differences between the
organization (i.e., KLPGA and LPGA) with regard to the number of followers,
likes, retweets, network sizes, and the total number of ties among Twitter stake-
holders. As indicated in Table 6.1, regarding the LPGA, the average number
of likes (M = 39.2) and retweets (M = 6.6) of each tweet, which could serve as
indicators of “interestingness”, is higher than the number of likes and retweets
the KLPGA garnered (0.4 for likes, 0.1 for retweets), meaning that followers
are more engaged with LPGA’s tweets, which they demonstrate by liking or
retweeting LPGA’s tweets.
The size of the social network (i.e., numbers of nodes) represents the prime
variable investigated in the analysis of a given online communication platform.
In this study, the total number of users mentioned (nodes) that is considered to
be a network size was 5 × 5 for the KLPGA and 368 × 368 for the LPGA, and
the total number of mentions (ties) among the tweets for both the KLPGA and
LPGA was 6 and 3,708, respectively (see Table 6.1). Only a few users and ties
were identified in the KLPGA network because the KLPGA uses its Twitter
webpage primarily as a way to share the results of KLPGA players’ tours and
not as a means of interacting with other stakeholders. On the other hand, the
LPGA focuses on interacting with online stakeholders. Specifically, the LPGA
mentioned at least one user in 2,399 of 3,162 tweets (76%).
To identify the KLPGA’s and LPGA’s primary online stakeholders, there were
11 categories for the coding of the data affiliated with usermentions. These categories

Table 6.1 N
 umbers of Tweets, Likes, Retweets, Network Size, and Total Ties in
KLPGA and LPGA Twitter Pages

Tweets Average Likes Average Retweet Network Size Total Ties

KLPGA 2,922 0.4 0.1 5×5 6

LPGA 3,162 39.2 6.6 368 × 368 3,708

Note: For KLPGA, the data were collected for over seven years, while for LPGA the data were
collected for four and a half months.The average score of likes and retweets is for each tweet.
116 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

Table 6.2 C
 ategorization of Online Stakeholders in the KLPGA and
LPGA Networks

Nodes Size
Categorization of
Stakeholders KLPGA (@korealpga) LPGA (@LPGA)

Women Golfer 0 147

Media Outlet 0 41
Golf Tournament 0 28
Sponsor 0 23
Public 6 23
Neither Sport nor Sponsor 0 23
Golf Course 0 21
Other Sports 0 20
Golf Association 0 16
Celebrity 0 15
Men Golfer 0 11
Total 6 368

Note: The KLPGA data were collected for over seven years, while the LPGA
data were collected for less than five months.
Nodes size = the number of users who had communication-based rela-
tionships with the KLPGA and LPGA in each group.

included women golfers (e.g., @LydiaKo), men golfers (e.g., @TigerWoods), golf
courses or clubs (e.g., @KingsbarnsGL), golf tournaments (e.g., @ROAD2LPGA),
sponsors or partnerships (e.g., @PNCBank), other sport players or organizations
(e.g., @serenawilliams), golf associations (e.g., LPGAGirlsGolf), celebrities (e.g.,
singers, actors, politicians), media-related stakeholders (e.g., @NBCSports), pub-
lic users (e.g., non-golfer and non-celebrities), and non-sport-related organizations
or sponsors (e.g., university). Table 6.2 indicates the numbers for each group of
Twitter users that represented LPGA stakeholders. When responding to public
user’s questions, the KLPGA barely interacted with six public users. On the other
hand, the LPGA was more inclined to mention different women professional
golfers (n = 147) than it was to mention users from other categories. The sec-
ond most mentioned group for the LPGA comprised media-related users (n = 41).
Compared to other groups, men professional golfers did not comprise a major user
group for the LPGA (n = 11).
For network analysis purposes, an adjacency matrix of 368 × 368 LPGA net-
work (numbers of cells = 135,424) was created using Python programming lan-
guage. The study suggests that high-degree nodes were closely associated with
the LPGA’s Twitter page. Within LPGA’s network, among other stakeholders, the
Golf Association engaged in the most interactions with other groups. For exam-
ple, as shown in Table 6.3, Golf Association engaged in a significant number of
interactions with women golfers (ties = 2,132), golf tournaments (ties = 701), and
media outlets (ties = 337).
The descriptive analysis (Tables 6.2 and 6.3) discussed here shows the
number of users per group and the number of ties among groups, but it does
 Social Media and Stakeholder Relationships 117

Table 6.3 Eleven Groups Users in LPGA’s Network

WG MG GC GT SP OS GA CE ME PU NS

WG 5 0 0 0 0 0 3 0 2 0 0
MG 0 0 0 0 0 0 0 0 0 0 0
GC 0 0 0 3 0 0 1 0 0 0 0
GT 21 0 6 0 1 3 15 4 6 1 1
SP 0 0 0 0 0 0 0 0 0 0 0
OS 4 0 0 5 0 2 7 2 0 0 0
GA 2132 31 49 701 64 38 57 39 337 86 24
CE 1 0 0 1 0 1 0 1 0 0 0
ME 13 0 0 5 0 0 10 0 7 0 0
PU 0 0 1 3 0 0 2 0 1 0 2
NS 4 0 1 1 0 0 3 0 0 0 1

Note: WG = Women Golfers, MG = Men Golfers, GC = Golf Course, GT = Golf Tournament, SP =

Sponsors, OS = Other Sport Players or Organizations, GA = Golf Association, CE = Celebrities,
ME = Media, PU = Public, NS = Non-Sport-related Organizations or Sponsors.

not make any specific suggestions regarding which categorical group tends
to be more central within the network, meaning that the groups have struc-
tural importance. This prompted the authors of the present study to conducts
social network analysis in order to identify any Twitter-based interactions that
occurred between the LPGA and its stakeholders and which groups of users
serve as the LPGA’s primary stakeholders groups. There are 11 categorical nodes
(Figure 6.1). Additionally, the number of ties a node has (i.e., degree) may be the
most straightforward measure of centrality. Because high-degree nodes are more
visible, they are generally considered important. Therefore, in the visualized
networks, the size of each node was adjusted by degree. In Figure 6.1, the net-
work pictured on the lower side was rearranged via a categorical attribute; this
permitted us to more readily recognize the LPGA’s primary stakeholders as well
as the strength of the interactions that occurred among the groups.
Within a network, a key property in terms of the position of a node is cen-
trality (Everett & Borgatti, 1999). Thus, via centrality data, the degree to which
certain nodes contribute to the network’s structure can be identified. As such, a
node’s centrality may be considered structurally important. In this study, degree
centrality was calculated using the number of users within each category in the
LPGA network. As this study’s findings illustrated, nodes of high-degree were
affiliated with the LPGA’s Twitter page (Figure 6.2). The LPGA frequently men-
tioned certain female professional golfers (e.g., Lexi Thompson, Stacy Lewis),
golf-tournament-related users (e.g., Ladies European Tour, Evian Championship),
and media outlets (e.g., Golf Channel) compared with other groups.
Social network software UCINET was employed to identify beta centrality val-
ues and to measure the total of each user’s direct and indirect potential influence
on all others (Borgatti et al., 2013). According to the results of the normalized
beta centrality scores, female professional golfers (e.g., @Lexi, @CKGolferChic),
golf tournaments (e.g., @RICOHWomensBrit, @CMEGroupLPGA), and media
118 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

Figures 6.1 N
 etworks for the LPGA. Networks Located in the Lower Side are Adjusted
by Groups (n = 368, ties = 3,708).

outlets (e.g., @GolfChannel, @TheAmyRogers) comprised most of the core actors

(Table 6.4), which illustrates the 30 core actors in the LPGA networks as based
on the beta centrality work of Bonacich (1987). This result did not come as a
surprise, considering the sizeable followings these users have.
Furthermore, with the use of NodeXL Pro, 9,654 tweets published on LPGA’s
official Twitter page between June 2 and September 26, 2018 were collected. The
criteria for this data set had to do with whether the tweets had an interaction with
the LPGA (@LPGA) via the Twitter interaction functions of mention (n = 6,564,
68%), retweet (n = 2,198, 23%), and reply (n = 892, 9%). Another interesting
 Social Media and Stakeholder Relationships 119

Figures 6.2 Networks for the LPGA by the Strength of Interactions (Degree).

point is that 18% of the collected tweets (n = 1,719) were written in other lan-
guages (e.g., Korean), which serves as an indicator of global fans’ interest in the
LPGA. As seen in Figure 6.3b, the LPGA engaged in direct interactions with
core users within this network, but their messages have been expanded via the
interactions among core users and subsequently via their interactions with users
positioned along the periphery (Figures 6.3c and 6.3d). For example, as shown in
Figure 6.3c, Angela Stanford, an American professional golfer currently compet-
ing on LPGA, is positioned in the center, but she has more Twitter interactions
with other users who are positioned toward the periphery. With regard to the
120 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

Table 6.4 3 0 Core Actors in the LPGA Twitter Networks Based on

Bonacich’s Beta Centrality

Users Bonacich’s Normalization Category

@Lexi 6.852 Women Golfer

@GolfChannel 6.779 Media
@RICOHWomensBrit 4.725 Golf Tournament
@CMEGroupLPGA 4.661 Golf Tournament
@EvianChamp 4.634 Golf Tournament
@IWITChamp 4.612 Golf Tournament
@CKGolferChic 4.14 Women Golfer
@NZWomensOpen 3.777 Golf Tournament
@BrookeHenderson 3.732 Women Golfer
@PortlandClassic 3.47 Women Golfer
@shanshanfengCHN 3.41 Women Golfer
@LydiaKo 3.192 Women Golfer
@TheAmyRogers 2.966 Media
@SWEET_IKKIM 2.906 Women Golfer
@Stacy_Lewis 2.546 Women Golfer
@ANordqvist 2.471 Women Golfer
@themichellewie 2.368 Women Golfer
@LPGA 2.225 Women Golfer
@1soyeonryu 2.189 Women Golfer
@ThePCreamer 2.031 Women Golfer
@Brittany1golf 1.831 Women Golfer
@ingeechun_dumbo 1.831 Women Golfer
@MarathonLPGA 1.791 Golf Tournament
@daniellekang 1.712 Women Golfer
@Gerinapiller 1.712 Women Golfer
@Cpwomensopen 1.577 Golf Tournament
@JuliInkster 1.552 Women Golfer
@NellyKorda 1.516 Women Golfer
@jutanugarn 1.482 Women Golfer
@LizetteSalas5 1.438 Women Golfer

PGA (see Figure 6.3d), although it is positioned along the periphery of the whole
network, it does have interactions with both users within the center as well as
those along the periphery. As such, network analysis regarding Twitter online
communication suggests that sport organizations can strategically interact with
groups located at a certain position in order to achieve desired outcomes.

Discussion and Conclusion

Despite the public’s significantly low levels of engagement with the KLPGA’s
tweets, the KLPGA has not made any noteworthy efforts to promote its events,
golfers, or campaigns using other stakeholders (e.g., golfers, celebrities, fans,
media). Further, the organization has not interacted with these stakeholders via
the Twitter platform. The LPGA, however, has numerous stakeholders, and it
 Social Media and Stakeholder Relationships 121

     

Figures 6.3 T
 witter Communication Networks for LPGA @LPGA and the Impact of
Distribution of Information via Angela Stanford (@angela_stanford) and
PGA (@PGA).
Note: Layout in square shape as follows: (a) upper left side, (b) upper right side, (c) lower left side,
and (d) lower right side.

has used its Twitter webpage to engage in a considerable number of interactions

with those stakeholders in the Twitter webpage; there are differences among
different types of users, though. That is, the LPGA had a much higher number
of interactions with LPGA players, golf tournament users, and media-related
users. These results can serve relevant organizations in that they can allow
them to better understand the ways in which social media users interact with
one another to spread information via the platform (Yang & Counts, 2010).
In order to address the research question at hand, this study evaluated network
structures, relationship characteristics, and organizational contexts pertinent to
the KLPGA and LPGA’s Twitter use. More specifically, the authors of this study
assessed the patterns of interactions that took place among the LPGA’s stake-
holders as they shared information. This was done so as to more fully under-
stand the networks that emerge as a result of the Twitter-based interactions that
122 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

happen between the LPGA and its stakeholders. Social network analysis was
employed here as a means of looking at these Twitter-based interactions and also
as a means of visually depicting the stakeholders’ networks. Further, to iden-
tify who was highly central within a given network, this study investigated the
number of nodes present in each network, the degrees of those nodes, and the
beta centrality of the social networks. If a node was found to be potentially
important, due perhaps to the large number of network ties that involved that
node, then the removal of that node might lead to a disconnect of the network.
As such, it is necessary to look for and identify nodes that could be central to
a network and thus could control the flow of information. Working under the
assumption that ties are where the flow of information exists, degree centrality
could be used as an index indicating a node’s exposure within a particular net-
work. Therefore, the position of a node could be considered the source of advan-
tageous and opportunistic possibilities. This study found that LPGA golfers, golf
tournaments, and media-related nodes were highly visible in the LPGA Twitter
networks and were, therefore, considered important. For example, with regard
to organizations, nodes with high degrees in organizational networks were often
the nodes insiders mentioned as being important to their respective groups.
However, per beta centrality scores, sports and sponsors (e.g., university), sport
PGA golfers, and other sport-related users were not highly positioned within the
LPGA Twitter network.
Overall, nodes are often referred to as being influential or as serving as gate-
keepers, and they are generally considered to possess autonomy, control, and
visibility while being engaged with others. Thus, sport organizations can use
centrality scores associated with social network analysis in order to more strate-
gically interact with highly centralized nodes that could affect the flow of infor-
mation. Sport organizations (e.g., the KLGPA) and relevant practitioners could
use the findings of this study to better employ social media platforms in order to
expand on their fan bases throughout the world. It is recommended that future
studies identify the reasons why the KLPGA uses social media only for infor-
mation sharing purposes and not to attract more fans to its events and athletes.
Additionally, it may be possible to examine the reasons behind the KLPGA’s
objectives as they pertain to social media use by conducting interviews with the
KLPGA’s social media managers.

References
Bonacich, P. (1987). Power and centrality: A family of measures. American Journal of
Sociology, 92, 1170–1182.
Borgatti, S. P., Everett, M. G., & Johnson, J. C. (2013). Analyzing social networks. Thousand
Oaks, CA: Sage.
Choi, S., & Park, H. W. (2014). An exploratory approach to a Twitter-based community
centered on a political goal in South Korea: Who organized it, what they shared, and
how they acted. New Media & Society, 16, 129–148.
 Social Media and Stakeholder Relationships 123

Eagleman, A. N. (2013). Acceptance, motivations, and usage of social media as a market-

ing communications tool amongst employees of sport national governing bodies. Sport
Management Review, 16, 488–497.
Everett, M. G., & Borgatti, S. P. (1999). The centrality of groups and classes. The Journal
of Mathematical Sociology, 23, 181–201.
Gibbs, C., O’Reilly, N., & Brunette, M. (2014). Professional team sport and Twitter:
Gratifications sought and obtained by followers. International Journal of Sport
Communication, 7, 188–213.
Han. E. (2013, October 13). Golf TV rights fees, the ceiling lot. The Korea Economic
Daily. Retrieved from http://land.hankyung.com/news/app/newsview.php?aid=
2013111244671
Han, E. (2014, March 26). 73.7% of women, 44.4% of men support. The Korea Economic
Daily. Retrieved from http://news.hankyung.com/article/2014032608941?nv=o
K League Information System. (n.d.). K League Attendance. Retrieved from http://
press.kleague.com/new/app_new/EtcRecord/AudienceAllSearch.asp?topCmsCd=
CM0001&parCmsCd=CM0069&CmsCd=CM0080
Kim, H. (2016, April 1). $1 Billion economic ripple effect, the economics of the KBO
(J. Yoon, Trans.). The Asia Business Daily. Retrieved from http://www.asiae.co.kr/news/
view.htm?idxno=2016040109580282985
Kim, K. (2015, May 8). The country’s three major land-based television stations’ expen-
sive K league broadcast rights (J. Yoon, Trans.). The Hankyoreh. Retrieved from http://
www.hani.co.kr/arti/society/media/690399.html
Kim, O. H. & Seo, H. J. (2018). A study on globalization process of Korean Women’s golf.
The Korean Journal of Physical Education, 57, 37–57.
Koo, S. K., Kim, J. H., & Won, Y. S. (2014). A study on physical, outward image of per-
ceptions and attitudes of the Korean professional women golfers. Korean Science & Art
Forum, 16, 25–35.
Korea Baseball Organization. (n.d.). KBO | Attendance History. Retrieved from https://
www.koreabaseball.com/History/Crowd/GraphYear.aspx
Korea Golf Association. (2017). 2017 Korean Golf Index. http://www.kgagolf.or.kr/Notice/
KoreaGolfView.aspx?p_BOARD_SE=3732&p_SEARCH_IF=&p_PAGE_NO=1#none
Korea Golf Course Business Association. (2017). The number of golf courses in Korea in
2016. Retrieved from http://www.kgba.co.kr/Media/Press_view.asp?id=108171&page=
1&searchItem=&keyword=&mode=23
Korea Ladies Professional Golf Association. (n.d.-a). KLPGA | Players. Retrieved from
https://klpga.co.kr/web/player/mover.do
Korea Ladies Professional Golf Association. (n.d.-b). KLPGA | Tour. Retrieved from
https://klpga.co.kr/web/tour/tourList.do
Korea Ladies Professional Golf Association. (n.d.-c). KLPGA History | Time Capsule.
Retrieved from https://www.klpga.co.kr/web/forward.do?rp=history/history
Korean Internet and Security Agency. (2016). 2016 survey on the Internet usage | Summary
report. Retrieved from http://isis.kisa.or.kr/board/index.jsp?pageId=060100&bbsId=
7&itemId=817&pageIndex=1
Kwon. H. (2017, November 16). Continuing popularity of KLPGA proven by the
number of spectators (J. Yoon, Trans.). The Yonhap News. Retrieved from http://www.
yonhapnews.co.kr/bulletin/2017/11/15/0200000000AKR20171115144900007.HTML
Ladies Professional Golf Association. (n.d.). LPGA. Retrieved from https://www.lpga.
com/
124 J. Yoon, W. (Eric) Jang, and P. M. Pedersen

Lebel, K., & Danylchuk, K. (2012). How tweet it is: A gendered analysis of professional
tennis players’ self-presentation on Twitter. International Journal of Sport Communication,
5, 461–480.
Lee, S. (2017, April 6). Expensive KBO’s broadcasting rights (J. Yoon Trans.). The Cider
Economics. Retrieved from http://www.cidermics.com/contents/detail/928
Lee, W. M., & Jung, H. (2016). The identity and appellation of Korean LPGA athletes in
newspaper. The Korean Journal of Physical Education, 55, 143–152.
Lee, C., Shin, J., & Hong, A. (2018). Does social media use really make people politically
polarized? Direct and indirect effects of social media use on political polarization in
South Korea. Telematics and Informatics, 35, 245–254.
Meyer, J. (2017, April 1). Players find a social media balance. LPGA. Retrieved from
https://www.lpga.com/news/2017-social-balance-tournament-weeks
Moon, S. (2006, July 18). Symbiotic relationship between golfers and sponsors (J. Yoon,
Trans.). The Weekly Kyunghyang. Retrieved from http://weekly.khan.co.kr/khnm.
html?artid=12402&mode=view
Naraine, M. L., & Parent, M. M. (2016). “Birds of a feather”: An institutional approach
to Canadian national sport organizations’ social-media use. International Journal of Sport
Communication, 9, 140–162.
Pfanner, E. (2012, July 1). Social media is the message for Olympics. The New York Times.
Retrieved from https://www.nytimes.com/2012/07/02/…/social-media-is-the-message-
for-olympics.html
Randall, M. (2012, July 4). Pak’s influence on game immeasurable. Golf Channel.
Retrieved from http://www.golfchannel.com/article/golftalkcentral/paks-influence-
game-immeasurable//
Rolex Ranking. (n.d.). Rankings. Retrieved from http://www.rolexrankings.com/en/
rankings/
Statista (n.d.). Sports & Outdoor: South Korea. Retrieved from https://www.statista.com/
outlook/259/125/sports-outdoor/south-korea
Thorpe, H. (2017). Action sports, social media, and new technologies: Towards a research
agenda. Communication & Sport, 5, 554–578.
Tracy, J. (2018, March 4). Changing the game: The impact of social media on golf.
Fairfield Hills Golf Course. Retrieved from http://www.fairfieldhillsgolfcourse.com/
changing-the-game-the-impact-of-social-media-on-golf/
White, A. (2018, February 20). Why the LPGA is investing in a social media tool to
help golfers build their digital brands. Front Office Sports. Retrieved from https://
frontofficesports.com/why-the-lpga-is-investing-in-a-social-media-tool-to-help-golfers-
build-their-digital-brands/
Yang, J., & Counts, S. (2010, May). Predicting the speed, scale, and range of information
diffusion in Twitter. Proceedings of the Fourth International AAAI Conference on Weblogs
and Social Media, 355–358. Retrieved from https://citeseerx.ist.psu.edu/viewdoc/
download?doi=10.1.1.481.291&rep=rep1&type=pdf
Yoo, S. K. (2012). The impact of global sport on local media: A case study of Korean
newspaper coverage of women’s golf. International Journal of Sport Communication, 5,
403–412.
Chapter 7

Influence of Micro-Celebrities
on the Formulation of Social
Media Marketing Strategies
Benjamin Mole, Paul Cook, and Ruth M Crabtree

Introduction
Social media can be considered an integral and deeply embedded part of daily
life that is relied upon by those who use it for various needs, for example, for daily
news updates, entertainment purposes, sport events, an information source, as
part of the user’s purchase decision-making process and to communicate with
other social media users around the world (Kapoor et al., 2018; Zhu & Chen,
2015). Social media currently has 3.96 billion active users, representing 51% of
the world’s population, who spend on average 2 hours and 24 minutes on social
media every day (Chaffey, 2020). In the modern era of marketing, social media
has become an essential part of marketing strategy, one that is being used by
the majority of companies worldwide (Vinerean, 2017). As a result of the huge
number of social media users, companies are enticed to exist in such a prosper-
ous online environment (Eroğlu & Bayraktar Köse, 2019; Balliauw et al., 2020).
Consequently, companies are investing significantly more resources into devel-
oping their social media marketing strategy, not only to increase their customer
relations online but also to aid in achieving their marketing and business objec-
tives (Zhu & Chen, 2015).
In recent years, one of the social media marketing strategies that compa-
nies have expressed an increased interest in is micro-celebrity endorsement
(Schouten et al., 2020). Micro-celebrities are not famous in the traditional
sense, instead they have strategically used social media to amass a large number
of followers, often on a global scale, becoming recognizable both online and
beyond the midst of their social media account, granting them a celebrity status
(Djafarova & Trofimenko, 2019; Eroğlu & Bayraktar Köse, 2019). Companies
are showing an increased interest in, and embracing the use of, micro-
celebrities to endorse their brand, products, and services (Schouten
et al., 2020). The allure of using micro-celebrity endorsements is that micro-
celebrities often have a social media following which far exceeds that of the com-
pany (Holt, 2016); therefore, the social media posts made by a micro-celebrity
can reach a larger audience. In addition, micro-celebrities develop a strong and
meaningful relationship with their followers and through an endorsement, the

DOI: 10.4324/9781003302476-7
126 B. Mole, P. Cook, and R. M. Crabtree

company believes that the micro-celebrity can increase its relationship with its
targeted consumers (Jun & Yi, 2020; Korontina & Jargalsaikhan, 2016). Although
there are numerous industries worldwide that have embraced the use of micro-
celebrities as a marketing tool, the fitness industry is arguably using micro-
celebrities to the greatest extent (Djafarova & Thompson, 2020).
Worldwide there has been a socio-cultural change with the global popula-
tion adopting a more active and healthier lifestyle, which has resulted in an
increased sales of fitness apparel, sport nutrition, and supplementation products
(Djafarova & Thompson, 2020). Subsequently, the fitness industry has become
one of the fastest-growing industries in the world, second only to the high
technology industry (Chekhovska, 2017). The fitness industry’s growth has
been so substantial that in 2019, it globally had an estimated value of
$100 billion (Biron, 2019). However, this industry is highly fragmented, rich
in competition, dynamic, and affected by the trends in both fitness and in
fashion (Korontina & Jargalsaikhan, 2016). Those who participate in fitness
activities have become increasingly more fashion conscious and are willing
to spend more money purchasing fitness apparel that is both fashionable and
practical (Djafarova & Thompson, 2020; Mintel, 2019).
The merging of fitness and fashion trends has led to the development of
fitness apparel known as Athleisure, which is clothing that can be worn for
social, casual, and fitness purposes (Euromonitor International, 2018). The
considerable growth of the fitness industry, alongside the increased demand
and sales of athleisure clothing, has substantially influenced the global sports-
wear market, with global sales exceeding $335.5 billion in 2019 (Euromonitor
International, 2019). In addition, arising from the demand in athleisure
clothing and the growth of the sportswear market, there has been a sudden growth
in the number of new brands emerging, alongside existing brands extending
their clothing range to include sportswear apparel, which has increased the
competitiveness of the sportswear market on a domestic and global scale
(Euromonitor International, 2018).
Celebrity endorsements have been a prominent marketing strategy that
have been extensively used by companies around the world since the late
19th century (Knoll & Matthes, 2017). As such, considerable research has
been conducted to investigate the effectiveness of celebrity endorsements, pre-
senting a surplus of evidence that highlights their benefits and substantiates
their use as a marketing strategy (Von Felbert & Breuer, 2020; Schouten et al.,
2020). There has been a noticeable increase in micro-celebrity endorsements
during the last decade, yet despite this, research investigating their effective-
ness is still within its infancy, thus their marketing value is scarce (Schouten
et al., 2020). Due to the lack of research conducted on this topic area, market-
ers may be apprehensive in integrating micro-celebrity endorsements as part
of their marketing strategy. Therefore, it is of importance to further develop
the understanding of the marketing value of micro-celebrity endorsements.
 Influence of Micro-Celebrities 127

Celebrity Endorsement
A celebrity can be defined as a well-known individual who receives a significant
amount of media and public attention, for example, actors, models, musicians,
and athletes (Schouten et al., 2020). Companies in various industries worldwide
have incorporated celebrities into their advertising strategy, using them to pro-
mote the company, their products, and services (Cuomo et al., 2019). One defini-
tion that encompasses all aspects of a celebrity endorsement has been provided by
Bergkvist and Zhou (2016), as a “celebrity endorsement is an agreement between
an individual who enjoys public recognition (a celebrity) and an entity (e.g., a
brand) to use the celebrity for the purpose of promoting the entity” (p. 644).
Celebrity endorsements are not a recent phenomenon, they are a well-
established advertising strategy that has been used for over 100 years with
one of the earliest examples dating back to 1893 (Louie et al., 2001). However,
the use of celebrity endorsements has seen a sharp rise in the last few dec-
ades. During the 1950s and 1960s, there was a considerable growth of the film
industry, alongside an increased household ownership of television sets (Erdogan,
1999). Subsequently, there was a surge in the number of film and televisions roles,
which by the 1970s, led to an abundance of celebrities available that companies
could select to endorse their products. The increased use of celebrity endorse-
ments was so significant that by the 1980s, one in five televisions advertisements
in the United States of America featured a celebrity endorsement (Erdogan,
1999). The persistent and extensive use of celebrity endorsements as an advertis-
ing strategy can still be seen in the present day, as celebrity endorsements con-
tinue to feature heavily on television advertisements on every continent around
the world (Bergkvist & Zhou, 2016). In addition, companies invest considerable
financial resources into this component of their advertising strategy (Upadhyay
& Singh, 2010), for example, Nike, a global sportswear brand, spends over
$6 billion per year on athlete endorsements (Enoch, 2020). This implies that
despite the financial costs involved, marketers consider celebrity endorsements to
be a valuable component of their advertising strategy (Upadhyay & Singh, 2010).
Arising from their extensive use and the financial cost involved, consider-
able research has been conducted to investigate the effectiveness of celebrity
endorsements as a marketing strategy. This is highlighted by the meta-analysis
conducted by Knoll and Matthes (2017), as their search results from three major
databases presented over 1,000 research articles relating to celebrity endorse-
ments. However, celebrity endorsement research is highly fragmented due to the
conflicting results from individual studies (Knoll & Matthes, 2017). This vari-
ation in research findings can potentially be explained, as individual research
studies often explore the research phenomenon from a different perspective or
using a different research approach, data collection or analysis method, or a
population demographic. Although the research findings are fragmented and
conflicting, there has been a number of literature reviews and meta-analyses
that have concluded that celebrity endorsements can be an effective marketing
128 B. Mole, P. Cook, and R. M. Crabtree

strategy, as they have shown to increase the consumers’ purchase intention,

increase brand and product awareness, and can improve brand loyalty, brand
recall and brand image (Bergkvist & Zhou, 2016; Knoll & Matthes, 2017). In
addition, Erdogan (1999) declares that celebrity endorsements are extremely
beneficial when entering foreign markets, as the barriers of entering the foreign
market of choice, such as language, culture, and time zones, can be broken down
due to the celebrity’s worldwide popularity.
Despite the aforementioned benefits, celebrity endorsements pose a number
of potential risks. For instance, the endorsed celebrity may be associated with
controversy, ill-behavior, or have accusations made against them, which not only
negatively impacts their image but also the image of the company and the prod-
uct they endorse (Upadhyay & Singh, 2010). The celebrity may sign an endorse-
ment deal with a number of companies, and therefore their image is associated
with an array of companies and products (Erdogan, 1999). This can tarnish the
credibility of the celebrity as consumers perceive them as only being motivated
by money (O’Mahony & Meenaghan, 1997). Furthermore, the popularity of the
celebrity may increase to such an extent that they overshadow the product they
endorse, or their popularity may decrease, which results in the consumer having
no interest in the celebrity and, therefore, no interest in the endorsed product
(Erdogan, 1999). Finally, celebrity endorsements incur substantial financial costs
with no guarantee on the return of investment (Upadhyay & Singh, 2010). Such
concerns signify that companies must take care when selecting a celebrity to
endorse their products. Fortunately, a number of models and theories have been
developed to aid in the celebrity selection process that can potentially increase
the marketing effectiveness of the celebrity endorsement. Namely, the Source
Credibility Model, the Source Attractiveness Model, the Match-Up Hypothesis,
the Meaning Transfer Model, and the Brand Alliance Concept.

Celebrity Endorsement Models and Theories

The Source Credibility Model implies that the effectiveness of the delivered
message depends on the credibility of the source, for example, the celebrity.
Credibility is a construct of two characteristics, expertise, and trustworthiness,
which the source must possess in order to be perceived as credible (Hovland
et al., 1953). Erdogan (1999) suggests that a credible source can influence
the attitudes, beliefs, and behaviors of the message’s recipient, hence, when
selecting a celebrity, credibility is of high importance. The Source Attractive
Model contends that the effectiveness of the message, in fact, depends on the
familiarity (awareness of the source), likability (affection of the source’s physi-
cal appearance and behavior), and similarity (resemblance between the source
and the receiver) of the source (McGuire, 1985; McCracken, 1989). Sources
that are familiar, likable, and similar to the consumer are perceived as attrac-
tive (McCracken, 1989). The source’s attractiveness not only encompasses any
physical attractiveness but any characteristic that the consumer perceives as
 Influence of Micro-Celebrities 129

attractive, such as intellect, personality, athletic prowess, and lifestyle. Erdogan

(1999) concluded that attractive endorsers can enhance the consumers’ attitudes
toward both product advertising and the brand. Yet, research findings are equiv-
ocal regarding their influence over consumers’ purchase intention, as mixed find-
ings have been reported (Erdogan, 1999; Kahle & Homer, 1985). The Source
Credibility Model and Source Attractiveness Model have been validated through
several research articles; however, McCracken (1989) has criticized these models,
as they imply the endorsement process is solely dependent on the celebrity and
places no consideration on either the endorsed product or the brand. This criti-
cism may potentially be rectified through the Match-Up Hypothesis.
The Match-Up Hypothesis suggests that the effectiveness of the endorsement
depends on the level of congruence between the endorser and the endorsed
product (Till & Busler, 2000). Multiple studies support this theory, whereby
a high level of congruence between the celebrity and the endorsed product
leads to a more effective endorsement, signified by consumers demonstrating
an improved perception, recall, and believability of a product advertisement
alongside an increased brand attitude and purchase intention (Schouten et al.,
2020; Till & Busler, 2000). In comparison, when the congruence between the
celebrity and the endorsed product is low, the celebrity is seen as a less credi-
ble source and the endorsement is considerably less effective (Schouten et al.,
2020). Therefore, companies should select a celebrity that has a high level of
congruence with their products and services, as this will increase the likeli-
hood of an effective endorsement. Unfortunately, according to McCracken
(1989), the Source Credibility Model, the Source Attractiveness Model, and
the Match-Up Hypothesis fail to incorporate the entirety of the endorsement
process, thus leading to the development of his own model that explains his
view of the endorsement process.
McCracken’s (1989) Meaning Transfer Model (MTM) consists of three stages
that describe his view of the celebrity endorsement process (Figure 7.1). Stage one
involves the creation of the image and personified the meaning of the celebrity
through the roles they have portrayed throughout their career. In stage two, the

Figure 7.1 Meaning Transfer Model. (Adapted from McCracken, 1989.)

130 B. Mole, P. Cook, and R. M. Crabtree

celebrity enters an endorsement deal, and their image and personified meaning
are then transferred onto the product they endorse. Stage three, the consumer
will purchase and consume the product to expect that the celebrity’s image and
personified meaning will be transferred unto them.
Though the MTM is one of the least researched models in the celebrity
endorsement context (Roy & Jain, 2017), previous studies have supported its
premise. For example, Langmeyer and Walker’s (1991) research findings showed
that symbolic meaning of the celebrity endorser, Cher, was transferred onto
the endorsed product, Scandinavian Health Spas. The sample, consisting of
51 undergraduate students, associated Cher with physical fitness, sexiness and
with an attractive physical appearance and by training at Scandinavian Health
Spas, they could obtain similar characteristics (Langmeyer & Walker, 1991).
Thus, supporting the premise of the MTM, as the meaning of the celebrity was
transferred onto the product, and by purchasing and consuming the product,
the consumer expects a similar meaning to be achieved.
However, the MTM model has also faced criticism as this model implies
that the transfer of meaning only occurs in one direction (Halonen-Knight &
Hurmerinta, 2010). Till’s (2001) findings support this criticism, as the image of
the celebrity was tarnished due to their endorsement of a negatively perceived
product. Therefore, this signifies that the transfer of meaning can occur in the
opposite direction, thus there is a two-way transfer of meaning from the celebrity
to the endorsed product and vice versa. This resulted in the authors proposing
the celebrity endorsement process should be considered as a Brand Alliance,
whereby the transfer of meaning exists on a pathway that incorporates the
celebrity endorser, the brand, and the endorsement action (Halonen-Knight &
Hurmerinta, 2010).
The models and theories that relate to celebrity endorsements highlight the
complexities involved within its process. Regardless, the use of celebrity endorse-
ments as an advertising strategy remains highly prevalent. Though celebrity
endorsements extensively feature on television advertisements in China and
South Korea, the extent to which they feature on television and print media
advertising across Europe and the USA is considerably lower than it was in the
late nineteenth century. Instead, in this modern era of advertising, companies
and celebrities are increasingly using the internet and social media as an advertis-
ing platform (Bergkvist & Zhou, 2016).

Social Media
Social media is a collective term for software-based technologies, such as an
application or website, that provides the user with a digital environment where
they can send and receive information or personal content and to interact with
other users worldwide (Appel et al., 2020). In the last decade, social media has
grown considerably, in 2010, there were an estimated 940 million social media
users (Moerdyck, 2010), which has since increased to approximately 3.96 billion
 Influence of Micro-Celebrities 131

in 2020 (Chaffey, 2020), equating to a 321% increase in the number of social

media users during this period. This growth is perhaps attributable to the rapid
developments in handheld technologies (Meng et al., 2015), as users have been
provided with a quick and easy access to the increased number of social media
platforms that have also emerged during this time period (Chaffey, 2020). Social
media encompasses a wide range of platforms, not only are the most popular
platforms such as Instagram, Twitter, and Facebook classed as social media,
weblogs, vlogging, and online forums are just some examples that also constitute
as social media. Appel et al. (2020) have a unique perspective of social media, as
in their view, social media has become less about the platform’s features and more
about how people are spending their time on social media, which thus far, has
primarily been the extensive sharing of information (Appel et al., 2020).
Traditionally, marketers have relied upon the 4 P’s Marketing Mix framework
(product, place, price, and promotion), developed by McCarthy and Perreault
(1964), whereby careful consideration and optimal decisions of each of the 4 P’s
would result in the creation and execution of an effective marketing strategy
(Johnson, 2015). Social media’s emergence disrupted this reliance. It introduced
several new influences which subsequently increased the complexity of the
marketer’s role, the marketing environment, and thus, the creation and execu-
tion of an effective marketing strategy (Johnson, 2015). However, Mangold and
Faulds (2009) argued that social media should be considered as a hybrid element
of the promotion mix (one of the 4 P’s of marketing), as this medium facili-
tates the combination of traditional marketing communication tools (a direct
engagement from the company to the consumer) with an amplified form of
word of mouth (consumers talking to one another). In addition, to develop and
implement an effective marketing strategy, social media should be integrated
alongside the traditional promotion mix, which consists of advertising, per-
sonal selling, public relations, direct marketing, and sales promotion (Mangold
& Faulds, 2009). This implies that the 4 P’s marketing mix framework is still
applicable in the modern era of marketing, marketers merely need to adapt and
integrate social media in order to develop an effective marketing strategy.
The initial interest that companies had in social media stemmed from several
developments, for instance, the declined response rate of e-mail, rapid tech-
nological developments, the declining use of traditional media, the increased
time consumers spend on social media, and the low cost of implementing a
social media marketing campaign (Gillin, 2007). In addition, research suggests
that marketing managers reported that the online presence of their competi-
tors and social media’s growth, popularity, cost-effective and viral nature are
motivational factors that have led to the integration of social media into their
marketing strategy (Tsimonis & Dimitriadis, 2014).
Social media has presented companies with a remarkable marketing oppor-
tunity, one that is being used by over 90% of companies worldwide (Bennet,
2013) and according to Vinerean (2017) is essential to any business strategy.
Social media provides companies with a marketing medium that can efficiently
132 B. Mole, P. Cook, and R. M. Crabtree

disseminate information worldwide, which can reach customers regardless of

the geographic, demographic, social, or political boundary while facilitating
a two-way communication channel between the company and customer in an
accurate and timely manner, all of which is incapable of traditional marketing
communications (Zahoor et al., 2016). As such, companies are increasing their
resources, financial or otherwise, toward developing their social media strategy
(Machado et al., 2020; Neti, 2011). According to Ebrahim (2019), in the forth-
coming years, up to 71% of a company’s marketing budget will be allocated
toward its social media strategy in an attempt to enhance its social media pres-
ence. As social media is being discussed within a marketing context, it is impor-
tant to address this term. Dahnil et al. (2014) define social media marketing as
“using the social media platforms and its related technologies and features to
help achieve marketing objectives in conjunction with other marketing com-
munication tools” (p. 120).
Many companies are integrating social media into their marketing strategy
to coincide with their marketing objectives. For example, marketing directors
and general directors of business-to-business companies in the United Kingdom
reported that they are using social media to attract new customers, increase
customer relations online, increase brand awareness, and to communicate their
brand online (Michaelidou et al., 2011). In comparison, Turner and Shah (2014)
alluded that business-to-consumer companies can leverage social media to acquire
new customers and increase their sales revenue, which is the primary goal of the
majority of business-to-consumer companies (Palumbo & Herbig, 2000).
Social media is regarded as one of the marketing phenomena of recent years,
and as such, there has been a growing body of research that has aimed to
address various aspects and to understand the effectiveness of social media market-
ing. Alalwan et al. (2017) reviewed 144 articles relating to social media marketing
and upon their analysis, highlighted that social media marketing can enhance
the company’s promotional marketing activity, improve customer loyalty, and
can increase brand awareness and the consumers’ purchase intention. Research
articles that were not included in this review have presented similar findings,
as improvements to the consumers’ brand, loyalty and engagement were also
reported as a result of social media marketing efforts (Jibril et al., 2019). Despite
the abundance of supporting evidence and overwhelming popularity, Zhu and
Chen (2015) argue that social media is an ineffective marketing strategy, as
the nature of social media is incompatible with the nature of marketing. They
suggest that companies are using social media to promote their products and
services to aid in achieving their business objectives, whereas people are using
social media to connect with other social media users, as a creative outlet, and
as a source of entertainment. Thus, people are not using social media to be
subjected to advertising messages, supporting their argument. However, 93% of
social media users think that a company should have a social media presence
and 85% believe that the company should directly engage with them through
social media (Michaelidou et al., 2011). This signifies that regardless of whether
 Influence of Micro-Celebrities 133

social media marketing is effective or not, it is important that a company estab-

lishes a social media presence, as consumers have this expectation.
Social media, once primarily used to interact and share personal content with
other users online, is becoming more frequently used as an information source
and is being incorporated into the consumers’ purchase decision-making process
(Appel et al., 2020). Social media users can search for, and read the reviews of
a brand, product, or service created by other social media users, thus obtaining
pre-purchase information (Paquette, 2013). Research suggests that consumers
have shown to trust online reviews created by unknown social media users to
a greater extent than traditional media advertisements created by the company.
Hence, social media can be considered an appealing information source, as the
user has easy access to countless reviews online. Subsequently, as social media
users are increasingly using this medium to search for and read reviews, this has
led to the development of what is known as electronic word of mouth (Cheung
& Thadani, 2012).
Word of mouth (WOM) is one of the oldest and most influential means of
information transmission (Wang & Fesenmaier, 2004) with a sizeable body of
marketing literature providing evidence that WOM is an effective marketing
tool (Zamil, 2011). Advances in technology, the emergence of the internet and
social media have changed the way in which information is transferred, thus
providing an environment for electronic word of mouth (eWOM) communica-
tions (Cheung & Thadani, 2012). Similar to WOM, eWOM communication
has shown to influence brand image, the consumers’ purchase intention, and
to increase product interest and awareness (Paquette, 2013). Although WOM
and eWOM involve the same theoretical processes (i.e., the transmission and
exchange of information), numerous aspects make eWOM unique. Electronic
word of mouth involves a multi-exchange of information that can be dissemi-
nated at unprecedented speeds around the world with this information remain-
ing accessible until deleted (Cheung & Thadani, 2012).
In comparison, in traditional WOM communication, information is usually
exchanged in a private conversation amongst a small group of people, making
it difficult to replicate the accurate transferring of the same information to par-
ties absent from the initial conversation (Cheung & Thadani, 2012). Although
eWOM is a powerful marketing force, companies have shown concern for
eWOM, as the company is unable to control the dissemination of information
in an online environment, and the power of shaping its brand image is no longer
with the marketing department, instead the power lies with the consumers’
online content and connections (Huete-Alcocer, 2017). For example, a negative
review created about the company’s product can be shared and made visible to
millions of potential customers, possibly impacting the consumers’ purchase
intention and their perceived image of the company. As such, companies must
find a solution to increase their control of the consumers’ eWOM, reducing the
risk of negative information being visible to millions of users online. The emer-
gence of social media disrupted the reliance placed upon traditional marketing
134 B. Mole, P. Cook, and R. M. Crabtree

strategies and has provided companies with a new avenue to promote their
products and services. However, resulting from the significant integration, the
market is now saturated with competitors using social media as a marketing
medium. Therefore, marketers need to find new means of promoting their com-
pany, products or services in the online environment. One potential solution
for a company to differentiate themselves from the competition and to increase
their control of the consumer’s eWOM is by using micro-celebrities.

Micro-Celebrities Overview
Social media’s substantial growth resulted in the creation of a new form of
celebrity, known as a micro-celebrity (Khamis et al., 2017). Traditionally, a
celebrity status was achieved by receiving significant public and media atten-
tion, often as a result for professional talent or a considerable accomplishment
(Schouten et al., 2020). Whereas, micro-celebrities have strategically used
social media to amass a large number of followers, often on a global scale, thus
becoming well-known and recognizable both online and beyond the midst of
social media, granting them a celebrity status (Djafarova & Trofimenko, 2019;
Eroğlu & Bayraktar Köse, 2019).
Albeit traditional celebrities have also amounted a considerable social media
following, micro-celebrities have built their entire careers online and without
social media they would not be known to the public, nor would their celebrity
status exist (Schouten et al., 2020).
Micro-celebrities exist in different fields, for example, fashion, beauty,
food, and fitness and they consider their followers as fans and strategically
share self-generated content on their social media accounts in an attempt to
increase their awareness, popularity, and their number of followers (Djafarova
& Thompson, 2020; Schouten et al., 2020). The content that micro-celebrities
share on their social media accounts primarily relates to the field in which
they operate (e.g., fitness) and often consists of product reviews that have no
marketing intention, leading to social media users perceiving micro-celebrities
as a credible source of information (Lima de Carvalho, 2017). Consequently,
micro-celebrities are able to build a highly regarded online reputation, allowing
them to be recognized as experts in their respective fields, which has permitted
certain micro-celebrities to turn their social media presence into a profession
(Schouten et al., 2020).
In comparison to traditional celebrities, social media users consider micro-
celebrities as more authentic and relatable, as micro-celebrities are seen as
normal people who have found fame online as opposed to being subject to the
star-making system (Neal, 2017). Furthermore, social media users can engage
directly with the micro-celebrity through social media and via face-to-face
meets (Su, 2017). Although traditional celebrities can also provide a similar
level of interaction with their followers, this interaction is pivotal for increasing
and maintaining the micro-celebrity’s popularity (Schouten et al., 2020). The
 Influence of Micro-Celebrities 135

interaction that the micro-celebrity has with their followers, through reply-
ing to messages or comments, has shown to increase the followers’ emotional
attachment with the micro-celebrity (Jun & Yi, 2020). Micro-celebrities also
share intimate details of their personal lives to a much greater extent than
traditional celebrities, which presents a persona that is less controlled and
managed (Marwick, 2015). A combination of the aforementioned has enabled
micro-celebrities to develop a strong and meaningful relationship with their
followers (Jun & Yi, 2020).
The rise of micro-celebrities has led to companies showing an increased inter-
est in and use of, micro-celebrities to endorse their products (Schouten et al.,
2020). The more followers a micro-celebrity has, the greater their perceived influ-
ence and thus, the greater their commercial value for a company (Djafarova &
Thompson, 2020). Companies select a micro-celebrity who exists in a similar field
and pay them to promote their products on their social media account and to
create content that highlights the product’s benefits while encouraging their fol-
lowers to purchase (Eroğlu & Bayraktar Köse, 2019). Micro-celebrities can create
viral content that can relay the company’s message, or advertise and promote the
company’s products directly in the view of thousands or millions of their follow-
ers that include the company’s target audience (Eroğlu & Bayraktar Köse, 2019).
Through the micro-celebrity endorsement deal, the company aims to transfer
the strong and meaningful relationship between the micro-celebrity and their
followers, onto the company (Djafarova & Thompson, 2020). Lima de Carvalho’s
(2017) findings support this expectation, as through a micro-celebrity endorse-
ment deal, the perceived credibility of the micro-celebrity was transferred onto
the company and vice versa. Thus, providing evidence that the MTM and the
Brand Alliance Concept apply to micro-celebrity endorsements.
Researchers suggest (Bruns & Jacobs, 2006; Neal, 2017) that micro-celebrity
endorsements are one of the hottest marketing trends that have the poten-
tial to be a highly profitable marketing strategy. Despite these claims, micro-
celebrity endorsements are primarily used by small-to-medium sized companies,
or companies exclusive to an online environment (Lim et al., 2017). Micro-
celebrity endorsements are an emerging marketing strategy, while there has been
an increased use of this strategy, their marketing value is not fully understood
(Schouten et al., 2020), perhaps defending the apprehension of large companies
implementing such a strategy.

Micro-Celebrities and the Fitness Industry

Micro-celebrity endorsements are being increasingly used in numerous indus-
tries; however, the fitness industry is arguably using micro-celebrities to
the greatest extent and effect (Djafarova & Thompson, 2020). As mentioned, the
fitness industry is in a state of considerable growth resulting from a worldwide
socio-cultural change whereby the global population is adopting a healthier and
fitter lifestyle (Djafarova & Thompson, 2020). This lifestyle change is perhaps
136 B. Mole, P. Cook, and R. M. Crabtree

attributable to the substantial fitness-related content on social media (Djafarova

& Thompson, 2020).
Social media has provided fitness-focused users with an outlet to share their
personal progress and to find information and guidance on workouts and nutri-
tion (Neal, 2017). As such, those with a lean and/or a muscular physique can
become an inspirational figurehead, amassing a large number of social media
followers and thus becoming a fitness micro-celebrity (Djafarova & Thompson,
2020). Fitness-focused users often follow the workouts and diets of fitness micro-
celebrities and purchase products based upon their recommendations with the
expectation that doing so will result in attaining a similar physique. Companies
within the fitness industry were quick to recognize this commercial opportu-
nity and incorporated fitness micro-celebrity endorsements into their marketing
strategy (Djafarova & Thompson, 2020). These companies pay fitness micro-
celebrities considerable sums, via one-off payments or sponsorships, to create
content that features their products (Neal, 2017).
Lima de Carvalho (2017) claims that social media users recognize fitness
micro-celebrities as a profession and have no negative attitudes toward their
use as a marketing strategy. Opposing these findings, Djafarova and Thompson
(2020) and Korontina and Jargalsaikhan (2016) reported that social media users
can develop negative attitudes and feelings toward the fitness micro-celebrity
when they excessively upload product advertising content. Comparable findings
have also been reported in other industries, whereby overexposure to product
advertisement negatively impacted the micro-celebrity’s credibility (Elli, 2017).
Korontina and Jargalsaikhan (2016) discovered that consumers developed a
negative attitude toward the micro-celebrity as they felt the micro-celebrity was
simply relaying the company’s message in exchange for payment. Hence, micro-
celebrities who excessively upload product advertising content might be perceived
as being motivated by money and only relay the company’s message as opposed
to giving their honest review. A similar notion was reported by O’Mahony
and Meenaghan (1997) with celebrity endorsements, therefore, micro-celebrities
must find an appropriate balance between uploading personal and advertising
content on their social media accounts to maintain their authenticity and credi-
bility while continuing their online profession.

Micro-Celebrity Endorsement and Purchase Intention

Calvo-Porral and Lévy-Mangin (2017) interpret purchase intention as the prob-
ability of the consumer purchasing a certain product or service in the present or
near future. The consumer purchase decision-making process (Figure 7.2) consists
of five stages that the consumer moves through once they realize they want to
purchase a product: need recognition, information search, evaluation of alter-
natives, purchase decision, and post-purchase decision (Solomon et al., 2019).
The time the consumer spends in each stage depends upon various factors such
as their level of interest, involved risk, and their previous purchasing experience
(Stankevich, 2017).
 Influence of Micro-Celebrities 137

Figure 7.2 Consumer Purchase Decision-Making Process. (Adapted from Solomon et al.,
2019.)

Although this model is widely used, it has, however faced criticism

(Stankevich, 2017). This model relies upon the consumer behaving rationally
and moving through each stage once they realize they want to purchase a prod-
uct; however, this is an inaccurate depiction of consumer’s purchase behavior
which can often be irrational, acting on impulsive without planning nor thought
(Solomon et al., 2019). Therefore, it can be assumed that this model only applies
to specific scenarios, when the consumer has identified and realized, they have
a need for a particular product or service. Korontina and Jargalsaikhan (2016)
reported that micro-celebrities influence the first four stages of the consum-
er’s purchase decision-making process. In addition, Djafarova and Trofimenko
(2019) also highlighted that micro-celebrities can cause the consumer to search
for information about the advertised product. However, micro-celebrities rarely
influence the purchase decision of the consumer, as they are perceived as not
a credible source of information. Contradicting Djafarova and Trofimenko
(2019), several research findings have shown that micro-celebrities can increase
the consumers’ purchase intention for an array of products in numerous indus-
tries (Chapple & Cownie, 2017; Schouten et al., 2020; Trivedi & Sama, 2020).
Many of these studies signified the importance of the Match-Up Hypothesis,
whereby a high congruence between the micro-celebrity and the product plays
a key role in influencing the consumers’ purchase intention. Therefore, it can
be theorized that due to the high congruence between a fitness micro-celebrity
and fitness-related products, consumers should demonstrate an increased intent
to purchase.
Research investigating the influence that micro-celebrities have on the con-
sumers’ purchase intention has primarily been conducted on youthful popula-
tions, as they account for the majority of social media users and are deemed as
being most influenced by micro-celebrity product advertising (Chun et al., 2018;
Djafarova & Thompson, 2020). However, there has been an increased use of
social media by those in an older age bracket, for instance those over the age of 50
(Pew Research Center, 2019), yet despite this, a finite number of studies have
investigated whether the micro-celebrity’s influence on the consumers’ purchase
intention differs between age groups. One such study that investigated whether
there was a relationship between age and an increased purchase intention was
conducted by Chun et al. (2018). Their findings reported that there was an
138 B. Mole, P. Cook, and R. M. Crabtree

insignificant relationship between age and an increased purchase intention of a

fashion item advertised by a micro-celebrity on Instagram.

Micro-Celebrity Endorsement Characteristics

Albeit emerging, there is limited research that has investigated whether the
widely accepted endorsement theories apply to micro-celebrity endorsements in
the online environment (Djafarova & Thompson, 2020; Djafarova & Trofimenko,
2019). Håkansson et al. (2020) discovered, upon review of the available litera-
ture relating to micro-celebrities, research has primarily been based around
four characteristics: trustworthiness, attractiveness, similarity, and authenticity.
Their research findings claimed that micro-celebrities should possess the afore-
mentioned characteristics in order to influence the consumers’ purchase inten-
tion. Lim et al. (2017) investigated the influence that micro-celebrities have over
the consumers’ attitude and purchase intention by using the Source Credibility
Model, the Source Attractiveness Model, the Match-Up Hypothesis, and the
MTM. They highlighted that the Source Attractiveness Model, the Match-Up
Hypothesis, and the MTM had a positive relationship and/or correlation with the
consumers’ attitude and purchase intention. No relationship was found between
the Source Credibility Model and consumers’ purchase intention and attitude,
indicating that the Source Credibility Model does not increase the effectiveness
of micro-celebrity endorsements.
Specific to the fitness industry, Djafarova and Thompson (2020) investigated
whether the Source Credibility Model, the Source Attractiveness Model, and
the Match-Up Hypothesis applied to fitness micro-celebrity endorsements in an
online environment. Their study conducted a content analysis on five Instagram
posts from ten popular fitness micro-celebrities, equaling 50 posts in total. By
using an analytic statistical software package, they discovered that fitness micro-
celebrities with a higher number of followers receive a higher level of engage-
ment (i.e., likes and comments), also a high quality of posted content influences
the fitness micro-celebrity’s perceived credibility. Respondents were also required
to state whether the characteristics included in the Source Credibility Model
and Source Attractiveness Model would increase the fitness micro-celebrity’s
perceived credibility. They were required to identify whether the characteristics
identified from the content analysis (high engagement, number of followers, and
quality of content) and whether experience, uniqueness, and charisma would
increase the fitness micro-celebrity’s perceived credibility. From these results,
a model can be formulated that identifies characteristics that will increase the
effectiveness of a micro-celebrity endorsement.
Trustworthiness, expertise, and the congruence between the product and
endorser were identified as being applicable to endorsement in both the online
and offline environment, whereas expertise, the number of followers, high
quality of content, uniqueness, and the number of likes and comments are newly
identified characteristics that are only applicable to endorsements in the online
 Influence of Micro-Celebrities 139

Figure 7.3 C
 haracteristics That Increase the Effectiveness of an Endorser in an Online
Environment. (Adapted from Djafarova and Thompson, 2020.)

environment. This research has provided marketers with a model, enabling

them to select a micro-celebrity that possesses the identified characteristics in
order to make the endorsement deal more effective. Furthermore, within this
research, the respondents were also required to state whether fitness micro-
celebrities have increased their purchase intention of fitness-related products.
Those who stated that fitness micro-celebrities had influenced their purchase
intention were asked to highlight which characteristics they perceived as
influencing their intention to the greatest extent. Knowledge, expertise, lean
and/or muscular physique, experience, and similarity to the respondent’s ideal
self-image were identified as the characteristics of a fitness micro-celebrity
that had the greatest effect on the respondents’ purchase intention. Although
not included in the developed model, these characteristics of a fitness micro-
celebrity are of importance to companies competing in the fitness industry,
as they provide additional characteristics that should theoretically lead to an
effective micro-celebrity endorser (Figure 7.3).

References
Alalwan, A. A., Rana, N. P., Dwivedi, Y. K., & Algharabat, R. (2017). Social media in
marketing: A review and analysis of the existing literature. Telematics and Informatics,
34(7), 1177–1190.
Appel, G., Grewal, L., Hadi, R., & Stephen, A. T. (2020). The future of social media
in marketing. Journal of the Academy of Marketing Science, 48(1), 79–95.
140 B. Mole, P. Cook, and R. M. Crabtree

Balliauw, M., Onghena, E. and Mulkens, S. (2020). Identifying factors affecting the value
of advertisements on football clubs’ and players’ social media: a discrete choice analy-
sis. International Journal of Sports Marketing and Sponsorship. Retrieved from https://doi.
org/10.1108/IJSMS-12-2019-0138
Bennet, S. (2013). 45 amazing social media facts, figures & statistics [INFOGRAPHIC].
Retrieved from https://www.adweek.com/digital/45-social-media-stats/
Bergkvist, L., & Zhou, K. Q. (2016). Celebrity endorsements: a literature review and
research agenda. International Journal of Advertising, 35(4), 642–663.
Biron, B. (2019). Fitness has exploded into a nearly $100 billion global industry as more
people become obsessed with their health. Retrieved from https://www.businessinsider.
com/fitness-has-exploded-into-a-nearly-100-billion-global-industry-2019-9?r=
US&IR=T
Bruns, A., & Jacobs, J. (2006). Uses of blogs. New York, NY: Peter Lang.
Calvo-Porral, C., & Lévy-Mangin, J. P. (2017). Store brands’ purchase intention:
Examining the role of perceived quality. European Research on Management and Business
Economics, 23(2), 90–95.
Chaffey, D. (2020). Global social media research summary July 2020. Retrieved from
https://www.smartinsights.com/social-media-marketing/social-media-strategy/new-
global-social-media-research/
Chapple, C., & Cownie, F. (2017). An investigation into viewers’ trust in and response
towards disclosed paid-for-endorsements by YouTube lifestyle vloggers. Journal of
Promotional Communications, 5(2).
Chekhovska, L. (2017). Fitness industry: state and prospects of development in the world’s
countries. Slobozhanskyi Herald of Science and Sport, 2 (58), 19–24.
Cheung, C. M., & Thadani, D. R. (2012). The impact of electronic word-of-mouth
communication: A literature analysis and integrative model. Decision Support Systems,
54(1), 461–470.
Chun, C. S., Lim, W. M., Tan, R. W., & Teh, E. W. (2018). Impact of social media influencer
on Instagram user purchase intention: The fashion industry (Doctoral dissertation, UTAR).
Cuomo, M. T., Foroudi, P., Tortora, D., Hussain, S., & Melewar, T. C. (2019). Celebrity
endorsement and the attitude towards luxury brands for sustainable consumption.
Sustainability, 11(23), 6791.
Dahnil, M. I., Marzuki, K. M., Langgat, J., & Fabeil, N. F. (2014). Factors influencing
SMEs adoption of social media marketing. Procedia-Social and Behavioral Sciences, 148,
119–126.
Djafarova, E., & Trofimenko, O. (2019). ‘Instafamous’ – credibility and self-presentation
of micro-celebrities on social media. Information, Communication & Society, 22(10),
1432–1446.
Djafarova, E., & Thompson, M. (2020). Exploring the credibility of fitness Instagram micro-
celebrities on young British males. Athens Journal of Mass Media and Communications,
6(2), 131–146.
Ebrahim, R. S. (2019). The role of trust in understanding the impact of social media mar-
keting on brand equity and brand loyalty. Journal of Relationship Marketing, 1–22.
Elli, D. M. (2017). The phenomenon and rise of influencer marketing and how it affects cus-
tomer opinion and helps or damages brands. Yayımlanmamış Yükseklisans Tezi, School of
Science & Technology (SST).
Enoch, BJ. (2020). Top 15 most influential Nike sponsored athletes on social. Retrieved
from https://opendorse.com/blog/top-nike-sponsored-athletes-on-social-media/
 Influence of Micro-Celebrities 141

Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing

Management, 15(4), 291–314.
Eroğlu, F., & Bayraktar Köse, E. (2019). Utilization of online influencers as an experien-
tial marketing tool: A case of Instagram micro-celebrities. Journal of International Social
Research, 12(63).
Euromonitor International. (2018). World Market for Sportswear. Retrieved from https://
www.portal.euromonitor.com/portal/analysis/tab
Euromonitor International. (2019) Sportswear in World. Retrieved from https://www.
portal.euromonitor.com/portal/StatisticsEvolution/index
Gillin, P. (2007). The new influencers: A marketer’s guide to the new social media. Fresno,
CA: Linden.
Håkansson, A., Jansson, E., & Kapteijn, N. (2020). The Mystery of social media
influencers influencing characteristics: An exploratory study on how social media
influencers characteristics influence consumer purchase intentions. Retrieved from
http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-96457
Halonen-Knight, E., & Hurmerinta, L. (2010). Who endorses whom? Meanings transfer
in celebrity endorsement. Journal of Product & Brand Management, 19(6), 452–460.
Holt, D. (2016). Branding in the age of social media. Retrieved from https://hbr.
org/2016/03/branding-in-the-age-of-social-media
Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion:
Psychological studies of opinion change. New Haven, CT: Yale University.
Huete-Alcocer, N. (2017). A literature review of word of mouth and electronic word of
mouth: Implications for consumer behaviour. Frontiers in Psychology, 8, 1256.
Jibril, A. B., Kwarteng, M. A., Chovancova, M., & Pilik, M. (2019). The impact of social
media on consumer-brand loyalty: A mediating role of online based-brand community.
Cogent Business & Management, 6(1), 1673640.
Johnson, N. (2015). The future of marketing: Strategies from 15 leading brands on how authen-
ticity, relevance, and transparency will help you survive the age of the customers. Upper
Saddle River, NJ: FT.
Jun, S., & Yi, J. (2020). What makes followers loyal? The role of influencer interactivity in
building influencer brand equity. Journal of Product & Brand Management, 29(6), 803–814.
Kahle, L. R., & Homer, P. M. (1985). Physical attractiveness of the celebrity endorser: A
social adaptation perspective. Journal of Consumer Research, 11(4), 954–961.
Kapoor, K. K., Tamilmani, K., Rana, N. P., Patil, P., Dwivedi, Y. K., & Nerur, S. (2018).
Advances in social media research: Past, present, and future. Information Systems
Frontiers, 20(3), 531–558.
Khamis, S., Ang, L., & Welling, R. (2017). Self-branding, ‘micro-celebrity’ and the rise of
social media influencers. Celebrity Studies, 8(2), 191–208.
Knoll, J., & Matthes, J. (2017). The effectiveness of celebrity endorsements: A meta-
analysis. Journal of the Academy of Marketing Science, 45(1), 55–75.
Korotina, A., & Jargalsaikhan, T. (2016). Attitude towards Instagram micro-celebrities
and their influence on consumers’ purchasing decisions. Retrieved from http://urn.kb.se/
resolve?urn=urn:nbn:se:hj:diva-31144
Langmeyer, L., & Walker, M. (1991). A first step to identify the meaning in celebrity endorsers.
Duluth, MN: ACR North American Advances.
Lim, X. J., Radzol, A. M., Cheah, J., & Wong, M. W. (2017). The impact of social media
influencers on purchase intention and the mediation effect of customer attitude. Asian
Journal of Business Research, 7(2), 19–36.
142 B. Mole, P. Cook, and R. M. Crabtree

Lima de Carvalho, I. (2017). Virtual sales floor: Consumer’s perception towards microcelebri-
ties sponsored content (Doctoral dissertation, Dublin, National College of Ireland).
Louie, T. A., Kulik, R. L., & Jacobson, R. (2001). When bad things happen to the endors-
ers of good products. Marketing Letters, 12(1), 13–23.
Machado, J.C., Martins, C.C., Ferreira, F.C., Silva, S.C., & Duarte, P.A. (2020). Motives
to engage with sports brands on Facebook and Instagram – The case of a Portuguese
football club. International Journal of Sports Marketing and Sponsorship, 21(2), 325–349.
Mangold, W. G., & Faulds, D. J. (2009). Social media: The new hybrid element of the
promotion mix. Business Horizons, 52(4), 357–365.
Marwick, A. (2015). You may know me from YouTube: (Micro-) celebrity on social
media. In P. D. Marshall & S. Redmond (Eds.), A companion to celebrity (pp. 334–350).
Chichester, England: John Wiley & Sons.
McCarthy, E. J., & Perreault Jr, W. D. (1964). Basic marketing. Homewood, IL: Richard
D. Irwin, 181.
McCracken, G. (1989). Who is the celebrity endorser? Cultural foundations of the
endorsement process. Journal of Consumer Research, 16(3), 310–321.
McGuire, W. J. (1985). Attitudes and attitude change. The Handbook of Social Psychology,
233–346.
Meng, M. D., Stavros, C., & Westberg, K. (2015). Engaging fans through social media:
implications for team identification. Sport, Business and Management: An International
Journal, 5(3), 199–217.
Michaelidou, N., Siamagka, N. T., & Christodoulides, G. (2011). Usage, barriers, and
measurement of social media marketing: An exploratory investigation of small and
medium B2B brands. Industrial Marketing Management, 40(7), 1153–1159.
Mintel. (2019). The activewear market grows more crowded as fitness studios become
competitors – 1st July 2019. Retrieved from https://reports.mintel.com/display/966140/
?fromSearch=%3Fargs%3D%252Fhomepages%252Fdefault%252F%26
freetext%3DFitness%26start%3Dredirect
Moerdyck, A. (2010). 940 million social media users in the world. Retrieved from https://
insites-consulting.com/940-million-social-media-users-in-the-world/
Neal, M. (2017). Instagram influencers: The effects of sponsorship on follower engage-
ment with fitness Instagram celebrities. Retrieved from https://scholarworks.rit.edu/
theses/9654
Neti, S. (2011). Social media and its role in marketing. International Journal of Enterprise
Computing and Business Systems, 1(2), 1–15.
O’Mahony, S., & Meenaghan, T. (1997). The impact of celebrity endorsements on con-
sumers. Irish Marketing Review, 10(2), 15.
Palumbo, F., & Herbig, P. (2000). The multicultural context of brand loyalty. European
Journal of Innovation Management, 3(3), 116–125.
Paquette, H. (2013). Social media as a marketing tool: a literature review. Retrieved from
http://digitalcommons.uri.edu/tmd_major_papers/2
Pew Research Center. (2019). Social Media Fact Sheet. Retrieved from https://www.
pewresearch.org/internet/fact-sheet/social-media/
Roy, S., & Jain, V. (2017). Exploring meaning transfer in celebrity endorsements: meas-
urement and validation. Asia-Pacific Journal of Business Administration, 9(2), 87–104.
Schouten, A. P., Janssen, L., & Verspaget, M. (2020). Celebrity vs. influencer endorse-
ments in advertising: the role of identification, credibility, and product-endorser fit.
International Journal of Advertising, 39(2), 258–281.
 Influence of Micro-Celebrities 143

Solomon, M. R., Askegaard, S., Hogg, M. K., & Bamossy, G. J. (2019). Consumer behav-
iour: A European perspective (7th ed.). New York, NY: Pearson.
Stankevich, A. (2017). Explaining the consumer decision-making process: Critical litera-
ture review. Journal of International Business Research and Marketing, 2(6), 7–14.
Su, A. (2017). Harnessing user-generated content through micro-celebrity endorse-
ments. Retrieved from http://www.brandba.se/blog/harnessing-user-generated-content-
through-micro-celebrity-endorsements
Till B. D. (2001). Managing athlete endorser image: The effect of endorsed product. Sport
Marketing Quarterly, 10(1), 35–42.
Till, B. D., & Busler, M. (2000). The match-up hypothesis: Physical attractiveness, exper-
tise, and the role of fit on brand attitude purchase intent and brand beliefs. Journal of
Advertising, 29(3), 1–13.
Trivedi, J., & Sama, R. (2020). The effect of influencer marketing on consumers’ brand
admiration and online purchase intentions: An emerging market perspective. Journal of
Internet Commerce, 19(1), 103–124.
Tsimonis, G., & Dimitriadis, S. (2014). Brand strategies in social media. Marketing
Intelligence & Planning, 32(3), 328–344.
Turner, J., & Shah, R. (2014). How to make money with social media: An insider’s guide to
using new and emerging media to grow your business (2nd ed.). New York, NY: Pearson.
Upadhyay, Y., & Singh, S. K. (2010). When sports celebrity doesn’t perform: How con-
sumers react to celebrity endorsement? Vision, 14(1–2), 67–78.
Vinerean, S. (2017). Importance of strategic social media marketing. Expert Journal of
Marketing, 5(1), 28–35.
Von Felbert, A., & Breuer, C. (2020). How the type of sports-related endorser influences
consumers’ purchase intentions? International Journal of Sports Marketing and Sponsorship.
Retrieved from https://doi.org/10.1108/IJSMS-01-2020-0009
Wang, Y., & Fesenmaier, D. R. (2004). Towards understanding members’ general partici-
pation in and active contribution to an online travel community. Tourism Management,
25(6), 709–722.
Zahoor, S. Z., Younis, M. I., Qureshi, I. H., & Khan, R. A. (2016). Impact of social media
marketing on brand equity. International Journal of Current Research, 8, 30046–30055.
Zamil, A. M. (2011). The impact of word of mouth (WOM) on the purchasing decision of
the Jordanian consumer. Research Journal of International Studies, 20(20), 24–29.
Zhu, Y. Q., & Chen, H. G. (2015). Social media and human need satisfaction: Implications
for social media marketing. Business Horizons, 58(3), 335–34.
Chapter 8

Examining Facebook Marketing

Practice
The Case of New Zealand Provincial Rugby
Jason Cole, Andrew J. Martin, and
Ashleigh-Jane Thompson

Introduction
Marketing literature documents a shift from a transactional paradigm to a
relationship paradigm (Gronross, 1994; 2005). This shifting school of thought,
along with social and technological drivers, has given rise to the rapid growth
and popularity of social media, particularly Facebook (Ballings et al., 2015).
Pronschinske et al. (2012) identify social media as one of the most salient devel-
opments in relationship marketing. As Ratten and Thompson (2021) argue,
the ideological foundation of the internet was to provide a social platform that
supported this relationship approach, which has now been further extended
with the advent of social media. Importantly, social media facilitates ongoing
two-way conversations between consumers and organizations, enabling frequent
anthropomorphized brand interaction as well as instant feedback between both
parties, characteristics that are central to relationship marketing. Arguably,
social media have become a defining feature of 21st-century communications.
Conceived in 2004, Facebook, in particular, has risen from relative obscurity to
become one of the most visited websites in the world.
While social media use has grown exponentially, so too has its influence.
Sport organizations were quick to capitalize on Facebook’s popularity particu-
larly with the introduction of brand pages in 2010. The trend is no different
particularly in New Zealand Rugby’s (NZR) National Provincial Championship
(NPC). With significantly more users than any other platform, it can be sug-
gested that Facebook is by far the most prominent social media channel for
rugby teams in New Zealand (Ballings et al., 2015). Within this relationship
paradigm, using this medium, sport managers can now listen and understand
fans’ needs, create “value adds” for fans to engage, and potentially translate into
the building, maintenance, and enhancement of relationships (Abeza et al.,
2019). Consequently, social media are widely accepted as important digital
marketing channels for many organizations, including sport teams (Abeza et al.,
2013; Ratten & Thompson, 2021). Indeed, one of the biggest changes within
the sport industry has been the introduction and uptake of social marketing.
These platforms have revolutionized traditional sport marketing practices be

DOI: 10.4324/9781003302476-8
 Examining Facebook Marketing Practice 145

enabling organizations to engage in more direct and real-time interaction with

their consumers.
Not surprisingly, with the popular growth and uptake across varying age
brackets, social media are featuring in a growing body of research. Achen et al.
(2017), Ballings et al. (2015), and Thompson et al. (2016) explain that because
of Facebook’s uptake and large user base, it has risen to become a major market-
ing channel for organizations and a central consideration in marketing strategy
for sport organizations. Facebook is widely acknowledged as the most promi-
nent social media platform in business-to-consumer businesses, including sport
(Abeza et al., 2019; Pronschinske et al., 2012). While social media are acknowl-
edged as important platforms for contemporary sport marketing practitioners,
scholars also acknowledge the benefits and challenges of social media in con-
temporary marketing practice. That said, recent research indicates a lack of
understanding and consistency in evaluating effectiveness within the context
of Facebook. Scholars have further acknowledged a need to move beyond simple
metrics as measures of performance.
The content and performance of actual Facebook posts by brand pages
underpins most of the research on social networking sites (Achen et al., 2017;
Achen et al., 2018; Bonsón & Ratkai, 2013; Pronschinske et al., 2012). Although
Bonsón and Ratkai (2013) quantified expected levels of likes, comments and
shares on a post there remains a lack of comparable context. Moreover, extant
literature fails to provide any contextual approaches that would allow practi-
tioners to accurately critique both positive and negative performance as well as
compare it against other sport team brand page posts on Facebook.

Review of Literature
Several scholars contend that since its initial conceptualization relationship
marketing has successfully been applied at multiple levels including organization-
to- consumer, business-to-business (B2B), and organization-to-employee.
Despite its multiple applications, the consumer remains at the center of dom-
inant conceptual models (Abeza et al., 2019; Bee & Kahle, 2006; Gronroos,
2005; Ratten & Thompson, 2021). While perhaps more commonly considered
within the business context, relationship marketing has emerged as an impor-
tant marketing paradigm within sport. Undeniably, professional sport operates
as a service-led industry where customer orientation is essential, and scholars
have argued that in most cases, sport involves some form of relationship market-
ing activity (Bee & Kahle, 2006). In the context of sport, Stavros et al. (2008)
interviewed several Australian professional sport franchises from Australian
Rules Football, Rugby League, Basketball, and Soccer. Their findings indicated
an understanding of relationship marketing’s contemporary prominence but
a lack of application acceptance and implementation (Stavros et al., 2008).
Contrasting findings were reported by Achen (2014) in the study of profession-
als representing United States professional sports leagues. Respondents viewed
146 J. Cole, A. J. Martin, and A.-J. Thompson

relationship marketing “as essential for the survival of professional sport teams
and used numerous tactics to build and maintain customer relationships” (p. 14).
Scholars argue that a successful approach to relationship marketing extends
beyond a short-term marketing activity (e.g., ticket and merchandise sales) and
instead considers a more long-term view that focuses on developing meaningful
relationships that are proactive, interactive, and enduring (Kim & Trial, 2011).
If organizations can successfully achieve this outcome, they are more likely to
benefit from enhanced loyalty, connection, and attachment of their fans. It is
not only organizations though that receive benefits from a relationship market-
ing approach. For consumers, it reduces choices, makes decisions cognitively
consistent and efficient, leading to increased satisfaction. Despite the distinct
mutual benefits of relationship marketing, it can also present unique challenges
(Gronroos, 2005; Stavros et al., 2008). Internally, relationship marketing is often
set up superficially as a strategic tool, as opposed to developing the long-term
established value of an organization’s culture. In a survey of 25 British profes-
sional football clubs, Adamson et al. (2005) established that relationship market-
ing failed at the majority of clubs who implemented it because there was a lack of
tangible commitment at both strategic and operational levels of the organization
coupled with under-resourcing. In addition, external consumers can present chal-
lenges as they become accustomed to a relationship marketing approach.
As with other industries, the rise of social media has not gone unnoticed
in sports. In line with the growing media-dominated sports consumer, research
shows that social media are vital tools for sports marketers, and, as several
scholars argue, social media provide an effective platform for sports organiza-
tions to execute a relationship marketing strategy (Abeza et al., 2019; Achen
et al., 2018; Pronschinske et al., 2012;). In particular, several sports-related
studies have illustrated the benefits Facebook provides to both professional,
amateur, and niche sport organizations (Abeza et al., 2013; Eagleman, 2013). For
example, after analyzing eight-participation-based running events in Ontario,
Canada, Abeza et al. (2013) outlined specific sport organization benefits of using
social media platforms such as Facebook. Specifically, Abeza et al. (2013) estab-
lished the benefits of a social media presence for relationship marketing. These
included better knowledge of fans, advanced consumer-organization interac-
tion, effective consumer engagement, and quicker evaluation of relationship
status with consumers. In a subsequent study, Abeza et al. (2014) discovered
that while Canadian National Sport Organizations (NSOs) utilized Facebook
as an effective information distribution channel, they struggled to generate
effective relationship dialogue.
Scholars have also argued that Facebook offers the potential to better under-
stand their fan bases (Abeza et al., 2013; Thompson et al., 2017). In relation-
ship marketing, an organization must know and understand the consumer as
an individual and Facebook present a realistic platform to achieve such knowl-
edge that would otherwise be unrealistic due to cost and time (Abeza et al.,
2013). The very nature of social media means that it allows for more regular
 Examining Facebook Marketing Practice 147

interaction with consumers to the point that marketing insight becomes more
relevant, timely, accurate, personal, and cost-effective for sport organizations. Off-
season brand interaction has also been highlighted as another Facebook benefit
for sport organizations (Abeza et al., 2013; Abeza & O’Reilly, 2014; Thompson
et al., 2017). Traditionally interaction with a sports brand only came as a result of
direct consumption of the core product. However, it is now widely accepted that
Facebook allows for real-time direct communication with fans allowing sports
organizations to communicate activities beyond the core product, allowing for
greater accumulation of both consumer groups and individuals’ information.
Consequently, scholars widely advocate the use of Facebook brand pages by
sports organizations, particularly professional teams, in the pursuit of marketing-
related objectives (Achen et al., 2018). Thus, necessitating practitioners to be
aware of both the benefits and challenges of managing a Facebook brand page.
Research has highlighted that benefits of Facebook as a marketing platform
include better understanding consumers, maintaining a brand presence, receiv-
ing instantaneous feedback, and providing the ability to assess the consumer
organization relationship status (Abeza et al., 2013; Achen et al., 2018; Thompson
et al., 2016). That said, research has shown that most platforms’ analytical setup,
including Facebook’s, make it hard to establish the information needed to con-
firm an interactive ongoing relationship has been established (Abeza et al., 2013;
Abeza & O’Reilly, 2014; Waters & Walden, 2015). Consequently, while several
benefits can be derived from social media use, it’s also important to acknowl-
edge that Facebook can also present challenges to sport marketing practitioners
alongside these benefits.
Despite a vast uptake of social media in the corporate industry, as well as
commercialized sport, it is still challenging to manage. Challenges of managing
a Facebook brand page have been shown to include credibility and reliability,
managing the impact of consumer-driven content, identifying offline consum-
ers, allocating sufficient resources, and guaranteeing the reach of brand content
(Thompson et al., 2016). Moreover, with large audience sizes, it is hard to manage
consumer posts to large organization-owned pages, and it is difficult to ensure that
fans will not set up their own pages to compensate for their passion or perceived
lack of relevant information being posted by the organization (Abeza et al., 2013).
The literature also outlines that management of such social media involves using
relational keywords in posts, ensuring content is entertaining, informative, per-
sonalized, and with prior permission and brand association (Kwok & Yu, 2013).
Lastly, the high frequency of posts that are overly commercialized should be
avoided (Cho & Cheon, 2004).
Previous research has noted the difficulties administrators often face in quanti-
fying or measuring performance. In Facebook-related literature, such performance
measures include reach, engagement, and components of engagement on brand page
posts (likes, comments, and shares). Reach as a measurement is in line with tra-
ditional digital advertising metrics, but it is also argued as being too transac-
tional (Ballings et al., 2015). Engagement is widely considered a more appropriate
148 J. Cole, A. J. Martin, and A.-J. Thompson

measurement of Facebook performance, as it is more relatable to a relationship

marketing approach. Also, winning percentage, as well as total network size, is
shown to influence the level of engagement for sports brand pages (Pronschinske
et al., 2012). Both total page reach and total page engagement are influenced by the
performance of posts with likes, comments, and shares currently positioned, by
some, as the most important metrics to consider (Bonson & Ratkai, 2013).
Despite the growing body of research on Facebook as a marketing platform,
much of the research focuses on measurements of reach, engagement, likes, com-
ments, and shares taken at face value. The literature lacks investigation into
quantifying performance in a manner that is contextual and comparable within
Facebook. That is, when referring to the context, previous research has limited
acknowledgment of a brand pages existence within, for example, a competition’s
popularity, win/loss records, and population of the home city. Consequently,
there is no indicative way identified in the literature, perhaps except for Achen
et al. (2017, 2018), to compare one page to another. Although current findings
indicate how to improve on these performance measurements, there is no com-
parable data or system to confirm that a brand page is performing at an effec-
tive level. Furthermore, negative performance indicators are yet to be analyzed.
Thus, we contend that there needs to be further investigation into establishing a
contextual evaluation of a sports brand’s performance on Facebook.
Moreover, much of the extant social media sport-related literature has largely
focused on specific geographical locales (e.g., North America and parts of Europe)
and those operating at the professional level (e.g., professional soccer leagues, North
American major leagues). Few studies have examined sports organizations operat-
ing at lower levels. This presents a noteworthy gap. While they are not, arguably,
as prominent and well-resourced as many of the organizations currently examined,
many other sport organizations would also occupy this space, and thus there is
relevance in understanding these issues in this context. This current research
builds on the literature to contribute toward the understanding of Facebook brand
page practice and measurement. For this research, four provincial rugby unions
were approached to examine the perceived challenges and benefits of Facebook
brand page administration. Specifically, this research sought to investigate:

1 What are the perceived challenges for Facebook brand page administrators?
2 Is total network size relatable to the performance of Facebook brand page
reach and engagement?
3 Are proportional measures relatable to the performance of Facebook brand
page reach and engagement?

Method
This research uses a prominent mixed-method case study design called concur-
rent triangulation (Creswell, 2013). Concurrent triangulation involves collecting
both qualitative and quantitative data simultaneously. Data is then compared to
 Examining Facebook Marketing Practice 149

establish confirmation, disconfirmation, validation, or corroboration (Creswell,

2013). Creswell (2013) asserts that utilizing both quantitative and qualitative
methods increases the level of triangulation that can be gained around a research
problem. Triangulation is an important characteristic of mixed-method case
study research (Yin, 2009).

Case Context
New Zealand’s NPC was originally formalized in 1976 and has gone through
several format changes in the last decade. In its current format, 14 provin-
cial teams compete across two divisions. For this study, a convenience sam-
ple was used to select the participating teams, with four National Provincial
Competition (NPC) teams making up the final research sample (henceforth
referred to as P1-P4). Since the introduction of Facebook brand pages in 2010,
all four have experienced mixed on-field success. Two teams were based in cit-
ies that also had Super Rugby franchises (a franchise-based professional rugby
competition featuring teams from Australia, Argentina, Japan, New Zealand,
and South Africa). In the context of New Zealand, provincial rugby unions are
affiliated to a specific Super Rugby team based on geographical proximity and,
in some cases, partial equity agreements. The NPC in its current format exists
as a second-tier competition in New Zealand behind Super Rugby. With signif-
icantly more users than any other platform, it can be suggested that Facebook
is by far the most prominent social media channel for rugby teams in New
Zealand. At the time of the study, each provincial team’s Facebook brand page
had more followers than any other social media account they held, including
Twitter and Instagram (see Table 8.1).

Data Collection
All four pages were run by a professional whose role was part of the market-
ing and/or public relations function. One of the co-authors was the brand page
administrator for one of the teams as part of their role as Brand Manager. They
were the sole person in charge of the teams’ Facebook page. The importance

Table 8.1 Participating Team Social Media Presence

Teams

P4 P2 P1 P3

Facebook 10,511 19,616 12,830 117,112

Instagram 3,597 3,545 5,617 7,315
Twitter 3,069 4,359 4,971 5,817
YouTube 53 73 25 1,595
150 J. Cole, A. J. Martin, and A.-J. Thompson

of including this perspective is supported by Wall (2006), who maintains that

rich data, personal experience, connection to a cultural context, and influ-
ence strengthen auto-ethnographic data collection. Questionnaires, contain-
ing a combination of closed and open-ended items, were completed by each
page administrator from the participating provincial unions (other than the lead
author) pre- and post-season. Closed items include simple “yes” and “no” answers,
as well as Likert Scales. Open-ended questions were used to allow the respond-
ents to reveal personal insight from their context, adding richness to the data
(Tashakkori & Teddlie, 2003). An auto-ethnographic journal was also recorded
by the lead author pre, during, and postseason. As Duncan (2004) outlines, the
salient difference between an auto-ethnography and an ethnography is that the
researcher is already an insider within the research context as opposed to trying
to become an insider. For this research, observations were recorded in a journal
format along with reflective writing and documentation of artifacts in the form
of screenshots.
Secondary data were sourced from Facebook’s analytical pages found in the
brand page administrator consoles (henceforth referred to as Facebook Insights
data; see Figure 8.1). Administrator access was granted to the lead researcher
by all participating page administrators to facilitate the collection of analytical
data. A constructed week sample was utilized to collect data from all four of the

Figure 8.1 Facebook Insights as Displayed in the Brand Page Administrator Console.
 Examining Facebook Marketing Practice 151

participating team’s Facebook brand pages. Stempel (1989) explains that con-
structed week samples see a variety of content on different days of the week as
cyclical, whereby days are selected at random to represent a seven-day week over
a defined period (Riffe et al., 2009). For this research, constructed week sampling
was used to represent a seven-day week during the pre, regular and post-season
periods. Both pre- and post-season data were randomly selected from a window
of six weeks. Regular season data were randomly selected across the full ten-week
window accounting for all ten rounds of the NPC.
Using Facebook’s Insights, data was exported into an Excel spreadsheet. For
each brand page, total reach, engagement, and components of engagement (e.g.,
likes, comments and shares) were recorded for each post. Additional insights
were also recorded for each post including date, day of the week, time posted, post
type, post characters, content description and when an applicable number of hash-
tags, hashtags used, post clicks, photo views, clicks to play, link clicks, video views,
and average view duration of videos. Data for paid content were also recorded, if
applicable, including budget spent, paid reach, total paid actions, paid photo clicks,
and paid links. Negative user actions were also captured and included actions to
hide a post, report as spam, hide all posts, and unlike the page. Negative actions were
considered as they could provide important insight into consumer-organization
relationship status as they may indicate disengagement and break down of the
relationship marketing process among a consumer group. To ensure consistency,
all posts were recorded at 10.00 pm (NZST) the same evening. In addition to
individual post data, total page performance for the week was recorded on the
Sunday of every week during the sample period at 10.00 pm. Data recorded also
included new page likes, total page likes, number of posts, total page reach, total
reach on posts, total engagement, likes on content, comments on content, shares on
content, post clicks, page ranking as listed by Facebook and general notes about
the content posted that week.

Data Analyses
A six-phase thematic analysis was used to analyze the data from both the ques-
tionnaires as well as the auto-ethnographic journal (Smith & Caddick, 2012).
These six phases included (1) immersion in the collected data, (2) collating and
coding the data, (3) identification of themes, (4) reviewing of themes, (5) refining
the themes, and (6) creation of the research paper itself. Analytical Facebook Data
from all four pages was also used to examine page reach and engagement before,
during, and following the NPC season. The Facebook Insights quantitative data
were analyzed by producing mean averages for each of the analytical measure-
ments recorded. This was done for both the data collected from the constructed
week sample for post-performance as well as the weekly total performance data
taken for overall page performance. Averages were then compared at face value to
determine how each of the participating brand pages performed when compared
to the other. For this research, “face value” refers to Facebook data that has been
152 J. Cole, A. J. Martin, and A.-J. Thompson

taken straight from Facebook Insights and not adjusted using any equations. The
mean averages were then represented as a percentage of average total page likes
for the entire collection period. The performance of the participating pages was
then compared again to see if, and how, rankings differed when considering the
proportional measurements.

Results and Discussion

Interestingly, prior experience among participants would suggest that social
media and Facebook are now considered an established technology. Positioning
Facebook as an established technology that solicits familiarity, awareness, and
experience among professionals indicates a clear shift forward from Eagleman’s
(2013) earlier findings, where Facebook was viewed as an accepted, but emerging,
communications tool among sports professionals. Participants identified innova-
tion, accuracy, being thick-skinned or open to feedback and having a sense of
humor as the preferred traits of a brand page administrator, which is, in part,
consistent with Gillan (2009).

Facebook Benefits
Participants in this study positioned Facebook as their most important social
media platform. Interestingly, this was largely based on audience size as opposed
to any specific features intrinsic to Facebook itself, which appears to be coun-
ter to others’ findings (Ballings et al., 2015; Thompson et al., 2016). Participants
identified three key benefits of maintaining a Facebook brand presence for their
respective teams and acknowledged that Facebook allowed for better knowledge
of consumers, effective consumer engagement, and quicker evaluation of relation-
ship status with consumers. These three benefits align with and are significant
components of relationship marketing practice (Abeza et al., 2013). Specifically,
for one team, Facebook has enabled the organization to gather knowledge on a
variety of consumer’s backgrounds as well encouraging the same consumers to
repeat “basking in reflected glory” (BIRG) actions more frequently.
For another team, Facebook allows them to engage more effectively with
consumers emphasizing being able to respond instantly. Within the context
of relationship marketing, this is important as reducing the response times
between enquiries has been found to increase the level of credibility and
therefore trust of an organization (Gronroos, 1994). Moreover, such instan-
taneous feedback allows for a two-way, and therefore relational, dialog with
consumers (Achen et al., 2017; 2018; Thompson et al., 2017). Another impor-
tant benefit of using Facebook is the ability to collect qualitative market insight
on the relational status with their consumers (Abeza et al., 2013), allowing
the teams to make more informed relationship marketing decisions. All par-
ticipants explained that Facebook allows them to establish the relationship
status they have with their consumer group at large. For example, through
 Examining Facebook Marketing Practice 153

Facebook interactions, two teams now understand that their online consumers
respond very negatively to the team losing and thus must ensure that their
content remains honest and accurate so not to create greater despondence fol-
lowing a loss.

Facebook Challenges
Four key challenges were identified that were the most salient among partic-
ipants: uncontrollable factors, maintaining credibility, guaranteeing reach,
and resource allocation. Abeza et al. (2013) position uncontrollable factors as
referring to user-managed pages. However, participants in the current study
more readily cited core product inconsistency (i.e., losing games) as the most
concerning factor that they could not control. Bee and Kahle (2006) suggest
that unpredictability would lead to higher levels of “cutting off reflective fail-
ure” (CORF) behavior, and therefore outward demonstrations of brand loyalty
would be less frequent.
Kotler et al. (2010) emphasize the ability of consumer-driven pages and content
to impact credibility, whether it be positive or negative. However, participants
in this study saw credibility as being influenced by their actions, or inaction,
more so than the action of consumers. Specifically, participants expressed that
commercially obligated content and posting relevant content in the off-season
presented key challenges in maintaining credibility among the followers. Given
the increasing importance of sponsorship revenue and the need to provide a
return on investment to sponsors, this challenge is one that may become further
exacerbated in the future and is something that page administrators will need to
manage carefully.
Participants described maintaining the credibility of content by ensuring it is
relevant to the core product as well as being honest about the core product per-
formance, especially in defeat. In contrast to the current findings, Aydin (2016)
positions credibility as largely being made up of the legitimacy of an information
source. Yaakop et al. (2013), however, describes credibility as being influenced
by the believability of content. Hence, utilizing honest relational language when
posting about a poor performance could be considered best practice (Aydin,
2016; Yaakop et al., 2013). Therefore, as the participant’s outline, reporting a
result is important; however, being honest about how the team performed on
route to the results is important to maintaining a credible relationship with
consumers.
All participants alluded to the difficulty in guaranteeing reach and engagement
on a post. Posts announcing winning results were suggested to consistently
be the most successful. These views appear to support Bee and Kahle (2006)
who suggested that successful teams were more likely to generate BIRG behavior
among fans via social media if content remained focused on the success of the
core product. Furthermore, P4 highlighted the challenge of remaining relevant
and generating interest in the competitions nine-month off-season. Gensler et al.
154 J. Cole, A. J. Martin, and A.-J. Thompson

(2013) describe brand relevancy as being strongest when communications are

easily associated with the core product. All participants reported some level of
spending on Facebook posts to help guarantee a desired level of engagement and
reach. However, the proportion of P4’s marketing budget allocated to social media
only equated to 0.0055%, while P3 only spent $10 (NZD) on boosting Facebook
content, despite overseeing the page with the highest number of followers. Despite
industry-wide recommendations for increased spending on social media (Abeza
et al., 2013), for these teams, spending remains minimal (i.e., less than 1% of each
teams’ marketing budget).
With the case of P4, a boosted post recorded in the sample only engaged six
more people than the average for purely organic content posted on the page.
Here, “organic” refers to Facebook activity that has not been paid for by the brand
page. Such findings appear to align with the sentiment of Lee et al. (2014), who
explain that while utilizing paid posts is encouraged by some, increasing spending
on Facebook paid media does not guarantee significant increases in reach and
engagement. While it is not clear what content will work for different organiza-
tions, it proposed that identifying and utilizing content that solicits high levels of
organic reach and engagement should be the priority before considering paid media
(Lee et al., 2014). Lastly, all participants highlighted a lack of time and human
resources as a challenge in maintaining their Facebook pages. Responding to com-
ments and direct messages to the page is acknowledged as the biggest consumer of
time for two of the pages. P2 also cited allocating their brand page administration
time to the P2 Super Rugby Team brand page as presenting a major challenge dur-
ing the post and off-season periods. This finding is consistent with prior studies
that argue the amount of time dedicated to Facebook brand page management is
the main resource-based challenge (Abeza et al., 2013).

Facebook Administration
The participating page managers were cognizant of the need to ensure that page
content is more conversational than transactional, and that content is interac-
tive. As argued earlier, credibility was described as important in practice, par-
ticularly when a team suffered a defeat. Participants sought credibility through
accurate and informative information that is delivered in a tone that suits the
context of the post and the organization itself. P1 and P2 described the need
to minimize commercial content frequency to avoid negatively impacting their
page credibility. In the context of the P4 page, commercially obligated content
was posted in September in line with an NZ rugby promotion for the NPC. This
content featured transactional words that included grab, prize, and competition.
However, the total engagement on the post was below the Page’s average for the
entire collection period. This finding is consistent with findings from Kwok and
Yu (2013), who found that relational language and structure of content could
improve engagement while transactional words such as “grab” and “competition”
could reduce performance.
 Examining Facebook Marketing Practice 155

Furthermore, participants describe that interactive content, such as photos

and videos, are the most successful to use on Facebook. The data from the
constructed week sample of Facebook posts appear to support the partic-
ipants’ position here. On average, posts that contained either a photo or a
video engaged more consumers than posts that did not. The data also show
that average negative post actions coincide with total network size, with P4
receiving the least and P3 the most on measures related to hide post and unlike
the page. However, with regard to hide all posts, P1 averaged more hide all post
actions (n = 0.58) than P2 (n = 0.50) despite having a smaller total network
size. P1 posted the most for the entire collection period (n = 53). This outcome
indicates that posting too often may hurt performance. Such findings appear
to align with Cho and Cheon’s (2004) study that found advertising avoidance
is more prominent when cluttering has occurred for the consumer. Thus, post-
ing too frequently on a Facebook page can clutter and interrupt a consumer’s
newsfeed, increasing the chance they will avoid or un-follow content from that
organization. Additionally, Facebook’s EdgeRank algorithm pushes content to
users who do not follow an organization’s page but are “friends” with someone
who does. Therefore, there is a high chance that cluttering could also have a
negative impact on other consumers’ opinions toward a brand due to the lack
of prior association (Unal et al., 2011). Acknowledging and recording nega-
tive engagement actions could lead to ensuring an ideal frequency of posts is
established for a brand page. Thus, page administrators must consider both
positive and negative measures.

Facebook Performance: Reach

P3 averaged the highest total reach for the entire collection period (n = 140,276),
while P2 averaged the least (n = 46,575). On average, the participating pages
reached 371.50% more people than their total network size. The performance of
the participating brand pages is exponentially more than what is suggested as
expected levels of reach in the literature. For example, Aydin (2016) noted that
brand pages could expect reach to continue to diminish moving forward. Both
Manson (2014) and DeMers (2015) explain that brand pages could expect to reach
16.00% of their total Facebook followers in 2012. However, as of 2015, averages
could be expected to be as low as 8.00% for pages with less than 50,000 followers
due to adjustments in Facebook’s EdgeRank algorithm (DeMers, 2015; Manson,
2014). Ballings et al. (2015) found that the greater network size, the greater reach
could be expected to be. Ballings et al. (2015) argued that reach was the most
relevant analytical measure to marketing practitioners using Facebook. Based
on this rationale, one would expect P2, who averaged the second most followers
(n = 18,081.22), to also average the second-highest total reach for the entire period.
The results, however, contrast with this, with P4 and P1 averaging better total
reach than P2 despite having fewer followers. It could be suggested that in con-
trast to the other participants, P2 is underperforming. Conversely, it could be
156 J. Cole, A. J. Martin, and A.-J. Thompson

suggested that both P4 and P1 are exceeding expectations. The contradictory

results of total reach taken at face value would suggest that it is not the most
accurate means by which to measure performance as well as make comparisons
of Facebook brand pages.

Facebook Performance: Engagement

Total weekly engagement varied across all three collection periods. P3 averaged
the most total page engagement for the entire collection period (n = 15,059.61).
The regular season saw the highest level of engagement for all participants. In
a similar fashion to reach, total network size did not necessarily correlate to
total engagement as P2 averaged the least engagement for the entire collection
period (n = 4,569.22). Conversely, winning percentage in the regular season
did not correlate with rankings by total engagement. During the regular-sea-
son, teams averaged the highest levels of total page engagement. During the
regular season on average 46.20%, more people were engaged than the pre-sea-
son and off-season periods. This point is important as Pronschinske et al.
(2012) found that the authenticity of brand page content was associated with
levels of engagement. Therefore, the regular season is an important time for
generating engagement.
In contrast to previous research (Ballings et al., 2015; Pronschinske et al.,
2012), as with reach, total network size did not correlate to total engagement for the
entire collection period. Of relevance is the fact that P2 has the second-highest
number of followers but averaged the least engagement for the entire collection
period. This contradicts Ballings et al.’s (2015) assertion that network size would
result in increased reach. While Pronschinske et al. (2012) found that higher lev-
els of engagement were significantly associated with the total network size, they also
found that frequency and quality of the content was an important factor in gen-
erating engagement and in turn network size. However, the discrepancy between
the findings of this current study and the literature would suggest that the actual
content plays a greater role and that network size alone cannot generate reach
and/or engagement (DeMers, 2015; Manson, 2014).

Facebook Performance: Posts

Post-performance rankings were consistent in terms of reach and engagement
when compared to total page reach and engagement. On average P3 had the most
reach (n = 17,740.81) and engagement (n = 554.71) across all thee collection periods
on individual posts. However, performance did vary across components of engage-
ment; likes, comments, shares. Once again, in contrast to Ballings et al. (2015),
reach on posts was not always directly relatable to total network size. Three of the
four pages’ average reach on individual posts did rank in the same manner as
their total network size (e.g., P3, P1, P4). However, P2 averaged the lowest reach on
individual posts despite having the second largest network size and best on-field
 Examining Facebook Marketing Practice 157

performance for the collection period. This would suggest that P2’s brand page
was either under-performing or, contrary to previous findings, post reach is not
significantly relatable to total network size or on-field performance (Ballings et al.,
2015; Pronschinske et al., 2012).
Similar to post-reach, post-engagement did not appear to be directly relatable to
network size. Again, three of the four participating pages’ engagement did rank
per their total network size (e.g., P3, P1, P4). However, once again, P2 averaged
the least engagement on posts for the entire collection period. This would once
again suggest P2 is either underperforming compared to other participating pages
in terms of engaging their audience and/or winning is not necessarily linked to
post engagement (Ballings et al., 2015; Pronschinske et al., 2012). On an average
post, engagement was broken down as 79.00% from likes, 10.16% from comments,
and 3.63% of shares, with the remaining 7.00% attributed to other post actions
(e.g., link clicks, photo views, and post clicks). Bonsón and Ratkai (2013) found that
brand pages could expect likes to make up 90.80% of engagement followed by com-
ments (n = 7.45%) followed by shares (n = 1.76%). With regards to comments, all
participants exceeded Bonsón and Ratkai’s (2013) findings by more than 10.00%,
with only P3 not exceeding the expected level of comments (n = 6.38%). All par-
ticipants exceeded Bonsón and Ratkai’s (2013) findings with regards to shares on
average by 1.87%, which is noteworthy given that Bonsón and Ratkai (2013) state
that shares demonstrate the strongest interaction with a brand. The level of per-
formance on comments and shares suggests that a two-way relationship does exist
between participating brand pages and their consumers.

Proportional Performance: Reach

Considering reach as a proportion of total network size generated a shift in page
performance rankings compared to face value performance. For the entire col-
lection period P4 reached an audience 549.00% the size of their total network,
followed by P1 (n = 478.00%), P2 (n = 260.00%) and P3 (n = 199.00%). During
the pre-season and post-season collection periods, P2 achieved the lowest
proportional ranking for reach. During the regular season and for the entire
collection period, average proportional rankings of page reach match total net-
work size rankings in reverse (e.g., P4, P1, P2, P3). Proportional rankings of
page reach being a reverse of total network size could be expected. Therefore,
as P4 had the smallest total network size and, in turn, the best proportional
ranking for reach during all three collection periods, it would initially suggest
that proportional ranking of reach is not necessarily an effective indicator of
performance.
However, during both the pre-season and post-season periods, P2 had the
lowest proportional reach. During the pre-season period, P2’s most recent core
product performance was the regular season campaign that saw them relegated
from the Premiership division with a 10.00% winning record. During the cur-
rent regular season, P2 produced a 60.00% winning record, the best of the
158 J. Cole, A. J. Martin, and A.-J. Thompson

participating teams. The post-season, however, saw them lose the Championship
final and, in turn, their chance to be promoted back to the Premiership divi-
sion. Furthermore, during the post and off-season periods, P2 shifts their time
and human resource focus to an SR brand page, reducing the amount of content
posted on the P2 NPC page. Hence, P2’s reduction in proportional reach dur-
ing the pre-season and post-season periods could be argued to reflect despond-
ence and disconnection with the core product among consumers during these
periods (Bee & Kahle, 2006). P2’s proportional reach performance during each
collection period is also reflected by the total reach data. This would once again
indicate that in contrast to Ballings et al. (2015), network size can correlate to
increased reach, but it does not guarantee reach. Therefore, representing reach
proportionally produces more comparable performance rankings by factoring in
total network size as a component of reach as opposed to the determining factor
that produces reach.

Proportional Performance: Engagement

Representing engagement as a proportion of total network size produced differ-
ent performance rankings across the pre-season, regular-season and post-season
collection periods compared to engagement taken at face value. During the
pre-season period P1 engaged the highest proportion of their total network (n =
77.00%) followed by P4 (n = 63.00%), P3 (n = 22.00%) and P2 (n = 13.00%).
During the regular season, P1 once again had the highest proportional engage-
ment (n = 75.00%) followed by P4 (n = 66.00%), P2 (n = 45.00%) and P3 (n =
29.00%). During the post-season P4 had the highest proportional engagement (n =
41.00%) followed by P1 (n = 38.00%), P3 (n = 10.00%) and P2 (n = 6.00%).
Pre-season proportional engagement potentially reflects the current relationship
between the brand pages and their respective consumer groups. Both P1 and P4
were approaching the upcoming season following division wins the previous year.
During the post-season collection period, P2’s proportional engagement decreased
by 39.00% from 45.00% to 6.00%, the largest decrease of all four participating
brand pages. This outcome could be a direct reflection of P2’s page administrator
shifting efforts to another brand page immediately following the NPC season’s
conclusion. Also notable is the engagement decreases experienced by the other
three participating pages, which ranged from 19.00% to 38.00%. The decreases
in proportional engagement would suggest that all the participating brand pages
struggle with disconnection from the core product during the post-season (Bee &
Kahle, 2006; Gensler et al., 2013; Waters & Walden, 2015).

Proportional Performance: Posts

P1 proportionally averaged the highest reach on individual posts (n = 45.99%)
collected in the constructed week sample. P2 recorded the lowest proportional
reach (n = 16.16%). P1 and P2’s proportional ranking contrasts with reach on
 Examining Facebook Marketing Practice 159

posts taken at face value. Furthermore, their ranking does not coincide with
their ranking according to total network size, unlike total proportional page reach.
Based on Ballings et al.’s (2015) findings, one might expect proportional rankings
to match total network size rankings in reverse (e.g., P4, P1, P2, P3). Deviations
from this could therefore represent above or below average performance when
compared to other brand pages. For example, P2 would have been expected to
have the third-highest proportional post-reach ranking given that they have the
second-largest total network size. P2 ranking fourth could be argued to be a sign of
the page underperforming within the context of the other participating brand
pages. Therefore, post-reach presented as a proportion of total network size could
be argued to be a comparable insight as to how brand pages are performing
against one and other.
P4 averaged the highest proportional engagement on individual posts (n =
1.91%), while P3 averaged the lowest (n = 0.79). Proportional post engagement
averages see the pages ranked according to total network size in reverse (e.g.,
P4, P1, P2, P3), suggesting it may be a predictable and therefore unnecessary
performance ranking. However, post comments and post shares represented as a
proportion of total network size are not comparable to page ranking according
to network size. P1 proportionally averaged the most comments per post (n =
0.17%). P2 proportionally averaged the most shares per post (n = 0.07%) followed
by P1 (n = 0.56%), P4 (n = 0.04%) and P3 (n = 0.02%). Considering the recom-
mendations of Bonsón and Ratkai (2013), it can be suggested that P1’s content
is better at producing higher levels of commitment from their consumer group
compared to the other brand pages. Conversely, by generating proportionally
more shares than the other brand pages, P2 could be argued to produce con-
tent that is more viral and that generates higher levels of BIRG among their
consumer group (Bonsón & Ratkai, 2013). These contrasting results indicate
a further challenge in determining how effective a page is performing over-
all. Therefore, effectiveness may ultimately be a result of what an organization
chooses to be its key performance indicator on Facebook.
P1 proportionally averaged the highest content-specific measures for post
clicks (n = 7.02%), clicks to play (n = 2.82%), link clicks (n = 1.24%) and video views
(n = 17.88%). P4 proportionally averaged the highest photo views per post (n =
2.29%). P2 proportionally averaged the lowest on each content-specific meas-
ure except link clicks which sees them rank third. These results would suggest
that P1’s content encourages more frequent interaction than other participating
pages (Yaakop et al., 2013). Interactivity is also considered to be a strong indi-
cator that content is effectively engaging consumers. As Yaakop et al. (2013)
note, high levels of interactivity lead to higher levels of commitment and trust
between the consumer and the organization, helping to ensure the consumer
is more open to commercialized content. Conversely, it could be suggested that
P4’s photos generate the highest level of interactivity among consumers while
P2’s content does not stimulate high levels of interaction at all (Yaakop et al.,
2013). Proportional representations of content-specific measures, therefore,
160 J. Cole, A. J. Martin, and A.-J. Thompson

provide context to brand comparable page performance as well as insights into

interactivity.
P1 proportionally averaged the highest levels of negative consumer actions
for both “hide post” (n = 0.0127%) and “hide all posts” (n = 0.005%). Only P3
and P2 had “unlike page” actions during the sample period. P1 posted the most
frequently of the four participating pages during the constructed week sample.
Cho and Cheon (2004) state that the high frequency of branded content can lead
to increased levels of avoidance among a consumer group. Hence, P1’s negative
action data could be the result of posting too frequently. Representing negative
consumer actions as a proportion of total likes provides context around P1’s data.
Should negative actions be represented at purely face value, P1 would have aver-
aged the third highest “hide post” actions and the second-highest “hide all posts”,
potentially leading one to conclude that post frequency was not an important
consideration.

References
Abeza, G., & O’Reilly, N. (2014). Social media platforms’ use in building stakeholder rela-
tionships; The case of national sport organisations. Journal of Applied Sport Management,
6(3), 103–126.
Abeza, G., O’Reilly, A., & Reid, I. (2013). Relationship marketing and social media in
sport. International Journal of Sport Communication, 6(1), 120–142.
Abeza, G., O’Reilly, N., & Seguin, B. (2019). Social media in relationship marketing:
The perspective of professional sport managers in the MLB, NBA, NFL, and NHL.
Communication and Sport, 7(1), 80–109.
Achen, R. (2014). Relationship marketing in United States professional sport: Attitudes,
opinions, and viewpoints of sport professionals. Global Sport Business Journal, 2(3),
14–27.
Achen, R. M., Kaczorowski, J., Horsmann, T., & Ketzler, A. (2018). Exploring off-season
content and interaction on Facebook: A comparison of U.S. professional sport leagues.
International Journal of Sport Communication, 11(3), 389–413.
Achen, R., Lebel, K., & Clavio, G. (2017). What customers want: Defining engagement
on social media in sport. Global Sport Business Journal, 5(3), 1–21.
Adamson, G., Jones, W., & Tapp, A. (2005). From CRM to FRM: Applying CRM in the
football industry. Database Marketing & Customer Strategy Management, 13(2), 156–172.
Aydin, G. (2016). Attitudes towards digital advertisements: Testing differences between
social media ads and mobile ads. International Journal of Research in Business Studies and
Management, 3(2), 1–11.
Ballings, M., Poel, D., & Bogaert, M. (2015). Social media optimization: Identifying an
optimal strategy for increasing network size on Facebook. Omega, 59(1), 15–25.
Bee, C., & Kahle, L. (2006). Relationship marketing in sports: A functional approach.
Sport Marketing Quarterly, 1(15), 102–110.
Bonsón, E., & Ratkai, M. (2013). A set of metrics to assess stakeholder engagement and
social legitimacy on a corporate Facebook page. Online Information Review, 37(5),
787–803.
 Examining Facebook Marketing Practice 161

Cho, C., & Cheon, J. (2004). Why do people avoid advertising on the internet? Journal of
Advertising, 33(4), 89–97.
Creswell, J. (2013). Research design: Qualitative, quantitative, and mixed methods approaches
(4th ed.). Thousand Oaks, CA: Sage.
DeMers, J. (2015). Why your organic Facebook reach is still falling – and what to do about
it. Retrieved on February 09, 2021, from http://www.forbes.com/forbes/welcome/
?toURL=http://www.forbes.com/sites/jaysondemers/2015/05/13/why-your-organic-
facebook-reach-is-still-falling-and-what-to-do-about-it/&refURL=&referrer=
#51edf9d02c84
Eagleman, A. (2013). Acceptance, motivations, and usage of social media as a market-
ing communications tool amongst employees of sport national governing bodies. Sport
Management Review, 16(4), 488–497.
Gensler, S., Völckner, F., Liu-Thompkins, Y., & Wiertz, C. (2013). Managing brands in
the social media environment. Journal of Interactive Marketing, 27(4), 242–256.
Gillan, P. (2009). Secrets of social media marketing: How to use online conversations and
customer communities to turbo-charge your business. Fresno, CA: Linden.
Gronroos, C. (1994). From marketing mix to relationship marketing: Towards a paradigm
shift in marketing. Management Decision, 32(2), 4–20.
Gronroos, C. (2005). The relationship marketing process: Communication, interaction,
dialogue, value. Journal of Business & Industrial Marketing, 19(2), 99–113.
Kim, Y., & Trail, G. (2011). A conceptual framework for understanding relationships
between sport consumers and sport organizations: A relationship quality approach.
Journal of Sport Management, 1(25), 57–69.
Kotler, P., Kartajaya, H., & Setiawan, H. (2010). Marketing 3.0: From products to customers
to the human spirit. Hoboken, NJ: Wiley & Sons.
Kwok, L., & Yu, B. (2013). Spreading social media messages on Facebook an analysis
of restaurant business-to-consumer communications. Cornell Hospitality Quarterly,
54(1), 84–94.
Lee, D., Hosanagar, K., & Nair, H. (2014). The effect of social media marketing content
on consumer engagement: Evidence from Facebook. Social Science Research Network,
1(1), 1–50.
Manson, M. (2014). Thinking social/value Facebook zero: Considering life after the demise
of organic reach. Retrieved on February 10, 2021, from https://social.ogilvy.com/
facebook-zero-considering-life-after-the-demise-of-organic-reach/
Pronschinske, M., Groza, M., & Walker, M. (2012). Attracting Facebook ‘fans’: The
importance of authenticity and engagement as a social networking strategy for profes-
sional sport teams. Sport Marketing Quarterly, 21(4), 221–231.
Ratten, V., & Thompson, A. (2021). Digital sport marketing. In P. Thaichon & V. Ratten
(Eds.), Developing digital marketing: Relationship perspectives (pp. 75–86). New York, NY:
Emerald.
Riffe, D., Aust, C., & Lacy, S. (2009). Effectiveness of random consecutive day and con-
structed week sampling. In K. Krippendorff & M. Bock (Eds.), The content analysis reader
(pp. 54–59). Thousand Oaks, CA: Sage.
Smith, B., & Caddick, N. (2012). Qualitative methods in sport: a concise overview for
guiding social scientific sport research. Asia Pacific Journal of Sport and Social Science,
1(1), 60–73.
162 J. Cole, A. J. Martin, and A.-J. Thompson

Stavros, C., Pope, N., & Winzar, H. (2008). Relationship marketing in Australian pro-
fessional sport: An extension of the Shani framework. Sport Marketing Quarterly, 17(3),
135–145.
Stempel, G. (1989). Content analysis. Research Methods in Mass Communication, 2(1),
124–136.
Tashakkori, A., & Teddlie, C. (2003). Handbook of mixed methods in social & behavioral
research. Thousand Oaks, CA: SAGE Publications.
Thompson, A., Martin, A., Gee, S., & Geurin, A. (2016). Fans’ perceptions of profes-
sional tennis events’ social media presence: Interaction, insight, and brand anthropo-
morphism. Communication & Sport, 1(1), 1–25.
Thompson, A. J., Martin, A. J., Gee, S., & Geurin, A. (2017). Fans’ perceptions of
professional tennis events’ social media presence: Interaction, insight and brand anthro-
pomorphism. Communication and Sport, 5(5), 579–603.
Unal, S., Erics, A., & Keser, E. (2011). Attitudes towards mobile advertising: A research
to determine the differences between the attitudes of youth and adults. Procedia – Social
and Behavioural Sciences, 24(1), 361–377.
Wall, S. (2006). An autoethnography on learning about autoethnography. International
Journal of Qualitative Methods, 5(2), 146–160.
Waters, R., & Walden, J. (2015). Charting fandom through social media communication:
A multi-league analysis of professional sports teams. PRism, 12(1), 1–18.
Yaakop, A., Anuar, M., & Omar, K. (2013). Like it or not: Issue of credibility in Facebook
advertising Asian Social Science, 9(3), 154–163.
Yin, R. (2009). Case study research: Design and methods (4th ed.). Thousand Oaks, CA:
Sage.
Chapter 9

Analyzing Sport Fan

Facebook Pages
Influences of Post Characteristics and
Brand Attributes on Fan Engagement
Shang-Chun Ma, Ching-Hung Chang, Shang-Ming Ma,
Wan-Chen Chen, and James J. Zhang

Introduction
Social media use is a global phenomenon, with 2.95 billion users worldwide in
2019 (Statista, 2020a). The most popular social networks usually display a large
number of user accounts or strong user engagement. For example, Facebook
was the first social network to surpass 1 billion monthly active users (Statista,
2020b). For professional sport organizations, key stakeholders such as sport
teams, leagues, athletes, events, and fans are using social media (Abeza et al.,
2017) for multiple purposes such as athlete endorsements, promotions, news
updates, relationship marketing (Abeza et al., 2019), brand promotion, and most
importantly, interacting and communicating with fan bases (Anagnostopoulos
et al., 2018; Filo et al., 2015).
When operating social media pages, information is shared in the form of posts,
and users can see and engage with posts by liking, commenting, and sharing.
Posting types such as status updates, photos, videos, and links are likely to catch
users’ attention (Luarn et al., 2015). Various post types can represent different
levels of media richness, commonly referred to as the vividness of online content
(Coyle & Thorson, 2001). Posting types can also facilitate varying levels of inter-
activity, referring to the degree to which users can influence the form and content
of the media environment (Liu & Shrum, 2002). These two post characteristics,
namely vividness, and interactivity, can potentially enhance online engagement
(de Vries et al., 2012). Indeed, research findings associated with sport settings
(Gyulai, 2016; Osokin, 2019) and non-sport settings (Cvijikj & Michahelles,
2013; de Vries et al., 2012; Luarn et al., 2015; Tafesse, 2015) have indicated that
different types of posts and post characteristics can influence the effectiveness of
fan engagement (i.e., likes, comments, and shares).
Professional sport teams have widely adopted social media platforms as an
effective branding technique to maintain and promote their brand and com-
municate with their fan bases (Anagnostopoulos et al., 2018). Researchers have
studied the use of diverse social media to build team brand associations, such as
Twitter (Parganas et al., 2015) and Instagram (Anagnostopoulos et al., 2018), and

DOI: 10.4324/9781003302476-9
164 S.-C. Ma et al.

have found that brand attributes as one major aspect of the brand association are
significantly related to the engagement of online followers. Despite such efforts,
an examination of fan engagement of social media with brand attributes without
considering post characteristics seems unsound because post characteristics are
equally vital as they can help capture social media users’ attention (Fennis &
Stroebe, 2010) and strengthen their level of engagement (Gunawan & Huarng,
2015; O’Reilly, 2015).
The purpose of this study was to examine the impact of post characteristics
(vividness and interactivity) and brand attributes (product-related and non-
product-related) would affect sport fans’ engagement (likes, comments, and shares)
with two professional baseball teams’ Facebook pages (Lamigo Monkeys and
Fubon Guardians) that were created and managed by the fans themselves. These
teams are a part of the Chinese Professional Baseball League (CPBL) of Taiwan.
Instead of focusing on team Facebook pages, the present study examined two
fan-created Facebook pages relating to CPBL. Given the fact that attendees of the
CPBL games have not increased significantly over the past ten years (CPBL, 2020),
an examination of fan-created social media may shed some light on a unique
perspective for the league and team managers to genuinely understand their
customers and develop effective marketing plans. Drawing theoretical insights
from post characteristics (Cvijikj & Michahelles, 2013; de Vries et al., 2012) and
sport team-branding models (Gladden & Funk, 2002; Bauer, Stokburger-Sauer, &
Exler, 2008), the present study aimed to examine (a) how different post charac-
teristics would affect fans’ engagement (i.e., like, comment, and share) with two
professional baseball teams’ Facebook pages that were created and managed by
their fans and (b) how brand attributes (product-related and non-product-related
attributes) would influence fans’ engagement with Facebook while controlling for
season periods (pre-season, regular season, and post-season).

Review of Literature

Post Characteristics: Vividness and Interactivity

Social media are considered an extension of online advertising. Thus, posts need
special characteristics or features that make them salient from the background
and able to capture customers’ attention (Fennis & Stroebe, 2010, p. 51). In par-
ticular, post vividness and post interactivity are two key characteristics affecting
consumer engagement with posts on social media (Luarn et al., 2015). Vividness
is the use of different ways to stimulate the senses, can be distinguished by the
richness of the sensory stimuli (Coyle & Thorson, 2001), and can lead to more
positive attitudes toward the brand post (de Vries et al., 2012). Media types, such
as text, images, and videos, represent different levels of media richness (Luarn
et al., 2015). The degree of vividness differs in the way that it stimulates various
senses (Coyle & Thorson, 2001). For example, a picture is more vivid than textual
information; similarly, a video is more vivid than a picture as the former enables
 Analyzing Sport Fan Facebook Pages 165

more sensory stimuli (i.e., hearing) than just sight. Research shows that highly
vivid banners can effectively stimulate consumers’ intention to click (Cho, 1999).
Recent studies have also revealed that consumers have a better response to vivid
posts on social media (de Vries et al., 2012; Osokin, 2019).
Interactivity is defined as “the degree to which two or more communication
parties can act on each other, on the communication medium, and on the mes-
sages and the degree to which such influences are synchronized” (Liu & Shrum,
2002, p. 54). It is also characterized by communications between companies and
customers, and among customers themselves (Goldfarb & Tucker, 2011). De Vires
et al. (2012) identified low (link to a website and vote), medium (call to act and
contest), and high (question and quiz) levels of interactivity of post characteristics
of brand fan pages on a social networking site. Liu et al. (2017) examined post
interactivity with low to high levels, including statements, links to a new page,
and encouraging fans to take action on the Facebook page of convenience stores
in Taiwan. Post characteristics differ in terms of the degree of interactivity and
can stimulate consumers to react. Numerous previous studies have displayed that
different types of posts, such as videos, can strengthen engagement (Gunawan &
Huarng, 2015; O’Reilly, 2015; Osokin, 2019). Many scholars have studied engage-
ment issues in social media use (Cvijikj & Michahelles, 2013; de Vries et al., 2012;
Kwok & Yu, 2013; Osokin, 2019; Santos et al., 2019), confirming that different
message characteristics can influence the effectiveness of brand posts. Hence,
this study extends previous theoretical perspectives on advertising from brand
fan pages.
Malhotra et al. (2013) recommended that posting photos can elicit the most
likes. Tafesse (2015) explored five top-selling motorized brands in the United
Kingdom and suggested that vivid brand posts have a critical positive effect on
the number of shares, whereas interactive brand posts have a critical negative
effect on the number of likes and shares. Gyulai (2016) found that supporters were
actively responsive to the posts with images and videos on the USA Wrestling
Facebook page. What is more, the study also indicated that sports fans tend to
interact with the posts, which encourages them to do something such as sharing
their opinions. Liu et al. (2017) examined the effects of social media character-
istics on customer engagement with the Facebook pages of convenience stores
in Taiwan and found that the higher the level of vividness and interactivity of a
brand post, the more the popular it would be. Osokin (2019) analyzed European
football and discovered that images and videos have a positive effect on the num-
ber of post likes. Santos et al. (2019) displayed that teams should create interac-
tive opportunities to endorse fans’ co-creation behaviors. As a result, we posit the
following hypotheses:

H1: The higher the level of vividness of a brand post, the more popular the
brand post.
H2: The higher the level of interactivity of a brand post, the more popular
the brand post.
166 S.-C. Ma et al.

Brand Attributes: Product-Related and Non-Product-Related

Brand attributes, as a part of the brand association categories (i.e., attributes,
benefits, and attitudes) (Keller, 1993), are the focus of the present study. Keller
(1993) distinguished two types of attributes: product-related and non-product-re-
lated. The former refers to those necessary for performing the primary product
or service function sought by consumers and the latter denotes external aspects
of the product or service that relates to its purchase or consumption, although it
does not relate directly to the product performance or service (Parganas et al.,
2015). In the sport context, Gladden and Funk (2002) firstly developed the Team
Association Model (TAM), consisting of 16 brand association aspects that were
divided into product-related attributes (i.e., success, star player, head coach, and
team management), non-product-related attributes (i.e., logo, stadium, tradition,
and product delivery), benefits, and attitudes. Bauer et al. (2008) modified the
TAM into product-related attributes (i.e., success, star player, head coach, and
team performance), non-product-related attributes (i.e., management, logo, sta-
dium, history and tradition, club culture and values, fans, sponsor, and regional
provenance), benefits, and attitudes. They applied the concept to the professional
football setting. Building upon Gladden and Funk (2002) and Bauer et al. (2008),
Parganas et al. (2015) examined brand attributes by using product-related (i.e.,
team success, star player, and head coach) and non-product-related (i.e., brand
mark, management, club’s history and tradition, club’s culture and values, events’
image, sponsor, fans, and stadium) aspects to analyze the Twitter page of a pro-
fessional football club. Recently, Anagnostopoulos et al. (2018) also adopted
Parganas et al.’s (2015) analytical model to study Instagram photos of two football
teams from the English Premier League.
Given the product-related and non-product-related attributes that are con-
ceptually and empirically important, which are connected with different social
media analyses in professional sports contexts, they were incorporated and
examined in the present study. Product-related attributes represent those fac-
tors that contribute to the performance of the team, such as success, star player,
head coach, and management (Gladden & Funk, 2001, 2002). In this study, we
adopt a classification modified from that of Gladden and Funk, defining success,
star player, head coach, team (members), and team performance (team play) as
product-related attributes. These attributes can influence consumption behavior
(Gladden & Funk, 2002; Kaynak et al., 2008). Bauer et al. (2008) demonstrated
that non-product-related attributes include such aspects as management, logo and
club colors, stadium, club history and tradition, club culture and values, fans,
sponsor or owner, and regional provenance.
Empirically, Watanabe et al. (2015) found that team performance is directly
significant to online engagement rates. Parganas et al. (2015) analyzed the use of
Twitter by a professional football organization to understand brand attributes and
their relationships to engagement features (reply, retweet, favorite)and found that
brand associations’ communication was transferred on social media with varying
 Analyzing Sport Fan Facebook Pages 167

Figure 9.1 The Hypothesized Conceptual Framework.

degrees of fan interactions. In particular, fans responded more often to product-

related attributes than to non-product-related attributes. Anagnostopoulos et al.
(2018) examined how professional team sport organizations used Instagram for
branding and influenced Instagram followers’ reactions. The results showed that
product-related attributes were used more often than non-product-related attrib-
utes and encouraged greater engagement from the followers. Santos et al. (2019)
revealed that sharing team standings with others on match days or using team
products may increase the level of fan engagement. As a result, we posit the fol-
lowing hypotheses (Figure 9.1):

H3: The more a brand post is about product-related attributes, the more
popular the brand post.
H4: The more a brand post is about non-product-related attributes, the less
popular the brand post.

Method
Three types of fan engagement behaviors were measured: numbers of likes, com-
ments, and shares (de Vries et al., 2012). The specific post characteristics that reflect
low, medium, and high vividness and interactivity are reported in Table 9.1. Post
vividness was categorized into three different levels (low, medium, and high) (de
Vries et al., 2012), and post interactivity was categorized into two levels (low and
high) (Cvijikj & Michahelles, 2013). No vividness and low interactivity were used
as base categories in the data analyses. Time period (Pargnas et al., 2015) and team
name (Anagnostopoulos et al., 2018) were two control variables. Post characteris-
tics and content were analyzed over the time periods selected based on the profes-
sional league’s schedule: pre-season (i.e., January and February), regular season (i.e.,
March to October), and post-season (i.e., November and December) (Li et al., 2019).
168 S.-C. Ma et al.

Table 9.1 Operationalization of Vivid and Interactive Post Characteristics

Variable Reference Coding

Vividness The richness of a brand post’s Low: photo (images)

formal features Medium: event (announces an upcoming
(de Vries et al., 2012) event on the page)
High: video (mainly from YouTube)
Interactivity The message content has Low: text (sentences without photo, link,
different levels of interactivity and video)
(Cvijikj & Michahelles, 2013) High: link (mainly to news sites or blogs)

Data Collection
There are four teams in the CPBL, namely the Lamigo Monkeys, the Fubon
Guardians, the Chinatrust Brothers, and the UniLions. However, the Chinatrust
Brothers and the UniLions groups are closed groups whose Facebook pages could
not be accessed. As a result, the participants in this study were only fans of the
Lamigo Monkeys and the Fubon Guardians. The data were retrieved on March 20,
2019. Online information is constantly updated and removed. Thus, we targeted
posts published between January 1 and December 31, 2017. There were a total
of 3,879 posts from 17,469 members of the Fubon Guardians’ group, and a total of
924 posts were collected from 9,448 members of the Lamigo Monkeys’ group.

Data Analyses
Content analysis was employed as the methodological approach. After reviewing
the literature on content analysis coding processes, we employed a content analy-
sis where one of the authors coded the initial social media content sample for anal-
ysis and then a second independent coder performed the same coding to check
for reliability (Potter & Levine-Donnerstein, 1999). The Kappa statistical analy-
sis method was adopted to evaluate the inter-coder consistency. Disagreements
were resolved through consensus. The Kappa coefficient in the current study is
0.94, which is above 0.8, indicating a high degree of reliability (Watson & Petrie,
2010). Multiple regression was used to examine the effect between post charac-
teristics, brand attributes, and fan engagement. We view the Lamigo Monkeys,
post-season, the text, combination (i.e., product and non-product attribute) as the
baseline in the dummy variable. The three dependent variables (number of likes,
number of comments, and number of shares) were skewed in their distributions
in the data. Thus, the log transformation was first conducted in the analyses.
Table 9.2 lists the categories and variables.

Results
A total of 924 posts were collected from 9,448 members of the Lamigo Monkeys’
group; the average number of likes per post was 97.03 (SD = 2.77), the average
number of comments per post was 13.05 (SD = 0.76), and the average number of
 Analyzing Sport Fan Facebook Pages 169

Table 9.2 List of Categories and Variables

Category Variables

Type of communication Photo, video, text, link, and event

Brand attributes Product-related: star player (individual athlete), sport
(Bauer et al., 2008; (rules and requirements, teams; not specific
Gladden & Funk, 2002) athletes), team/game (focus on results from team
sports); team/nongame (about the national teams
but no results)
Non-product-related: other people (fans; no star
player), sport issue (tickets, official schedule, logo,
clothes colors, stadium, sponsor or owner, location
of team)
Combination: both product- and non-product factors.

shares per post was 0.5 (SD = 0.04). There was a total of 3,879 posts from 17,469
members of the Fubon Guardians’ group; the average number of likes per post
was 225.25 (SD = 3.20), the average number of comments per post was 30.38
(SD = 1.68), and the average number of shares per post was 0.93 (SD = 0.03). In
the Lamigo Monkeys’ group, on average, about 58.77% of the posts contained
vivid characteristics, and about 41.23% contained interactive characteristics. In
the Fubon Guardians’ group, about 35.96% of the posts contained vivid char-
acteristics, and about 64.04% contained interactive characteristics (Table 9.3).
Table 9.4 presents the communication types over time periods. Specifically, the
photo was the most popular among vivid fan post characteristics, and the link
was the most popular among interactive fan post characteristics. The event of
communication type was less used among vivid fan post characteristics. Because
these characteristics did not show much variation, we decided to exclude them
from further analysis. The two groups had a lower utilization of Facebook from

Table 9.3 Descriptive Statistics of Explanatory Variables

Variable Coding No. of Post Percentage (%)

Lamigo Monkeys
Vividness Low: photo 395 42.75
Medium: event 1 0.11
High: video 147 15.91
Interactivity Low: text 114 12.34
High: link 267 28.90
Total 924 100
Fubon Guardians
Vividness Low: photo 837 21.58
Medium: event 6 0.15
High: video 552 14.23
Interactivity Low: text 362 9.33
High: link 2122 54.70
Total 3879 100
170 S.-C. Ma et al.

Table 9.4 Post Type Over Time Periods

Time period Photo Video Text Link Event

Lamigo Monkeys
Pre-season 10 1 1 21 0
Regular season 79 23 11 29 0
Post-season 51 22 17 26 0
Total 140 46 29 76 0
Fubon Guardians
Pre-season 30 11 2 158 0
Regular season 86 73 55 223 0
Post-season 69 32 14 176 1
Total 185 116 71 557 1

December to February. Post characteristics and content were analyzed over the
time periods selected based on the professional league’s schedule: pre-season (i.e.,
January and February), regular season (i.e., March to October), and post-season
(i.e., November and December).
Table 9.5 displays that there was no difference between post types and teams
in the pre-season period (χ2 = 6.00, p > .05). Both of them utilized link posts.
During the regular season, the chi-square test revealed a significant difference
between post types and teams (χ2 = 73.90, p < .01). Lamigo Monkeys’ followers
preferred photos (55.6%), followed by links (20.4%) and videos (16.2%). Fubon
Guardians’ followers prefer links (51%), followed by photos (19.7%) and videos
(16.7%) (Table 9.6). During the post-season period, post types were significantly
related to teams (χ2 = 50.28, p < .01). Lamigo Monkeys’ followers preferred photos
(44%), followed by links (22.4%) and videos (14.7%). Fubon Guardians’ followers
preferred links (60.5%), followed by photos (23.7%) and videos (11%) (Table 9.6).

Table 9.5 Chi-Square Test Results of Time Period by Team

Test Variable χ2 Df p

Time period
Pearson Chi-Square Pre-season 6.00 3 .11
Regular season 73.90 3 .00
Post season 50.28 3 .00
Pre-season .015 — .02
Goodman and Kruskal’s Tau Regular season .061 — .00
Post season .064 — .00
Team
Pearson Chi-Square Lamigo Monkeys 33.427 6 .00
Fubon Guardians 65.424 6 .00
Goodman and Kruskal’s Tau Lamigo Monkeys .044 — .00
Fubon Guardians .029 — .00
 Analyzing Sport Fan Facebook Pages 171

Table 9.6 The Results of Team*Post Type*Time Period

Time Period Team Photo Video Text link

Pre-season LM 30.3% n<5 n<5 63.6%

FG 14.9% 5.5% n<5 78.6%
Regular season LM 55.6% 16.2% 7.7% 20.4%
FG 19.7% 16.7% 12.6% 51.0%
Post-season LM 44.0% 19.0% 14.7% 22.4%
FG 23.7% 11.0% 4.8% 60.5%

Note: LM: Lamigo Monkeys; FG: Fubon Guardians; % within the team.

When using time period as the segmentation variable, the post type was signifi-
cantly related to the team in the pre-season (Tau = .015, p < .05), the regular-season
(Tau = .061, p < .01) and the post-season (Tau = .064, p < .01) periods. In terms of
time period, the strength of association between post type and team in order was
post-season, regular season, pre-season. When using the team as the segmentation
variable, post type was significantly related to time periods for Lamigo Monkeys
(Tau = .044, p < .01) and Fubon Guardians (Tau = .029, p < .01). In terms of the
team, Lamigo Monkeys had a higher relation to post type and time (Table 9.5).
In summary, links were the predominant communication type used in the
pre-season for both fan groups, whereas Lamigo Monkeys’ fans utilized photos,
and Fubon Guardians’ fans utilized links the most during the regular season and
post-season.
Table 9.7 displays the coding results by category of Facebook posts. Photos are
the prevalent communication type used when discussing team/game and sport

Table 9.7 Brand Attributes by Post Type (by Number)

Product-Related Non-Product-Related

Team/ Team/ Other Sport

Post Type Star Player Sport Game Non-Game People Issues

Lamigo Monkeys
Photo 35 13 54 11 9 13
Event 0 0 0 0 0 0
Video 13 1 22 1 7 1
Text 2 7 13 0 0 7
Link 48 5 42 1 4 5
Total 98 26 131 13 20 26
Fubon Guardians
Photo 86 6 63 12 26 6
Event 0 0 1 0 0 0
Video 54 2 35 5 27 2
Text 18 9 30 6 8 7
Link 354 45 248 10 95 41
Total 512 62 377 33 156 56
172 S.-C. Ma et al.

issues in the Lamigo Monkeys’ group. Links are the main communication type
used when referring to a star player in the Lamigo Monkeys’ group. Nonetheless,
links are the predominant communication type used when discussing star players,
team/games, and other people in the Fubon Guardians’ group.
The largest number of posts was published in October in the Lamigo Monkeys’
group when the Taiwan Series was held, whereas the largest number of posts
was published in May in the Fubon Guardians’ group. These findings further
indicate that the post content was related to brand star players since spring train-
ing was held during the pre-season in both groups. During the regular season
(i.e., October) and post-season (i.e., November), the most frequently discussed
issue was teams/games in the Lamigo Monkeys’ group because that was when the
Taiwan Series was held. The Lamigo Monkeys had 74 wins, 41 losses, and their
winning percentage was .655, whereas Chinatrust Brothers had 53 wins, 64 losses,
and their winning percentage was .453 for the whole year. As a result, the Lamigo
Monkeys were the champion of the Taiwan Series. On the other hand, the most
frequently discussed issue in the regular season (i.e., May) was star players in
the Fubon Guardians’ group because it was time to vote for the All-Star Game.
During the post-season period (i.e., November), fans were concerned about star
players and teams/games (i.e., Asia Winter Baseball League and Asia Professional
Baseball Championship) (Table 9.8).
Figure 9.2 shows that March, May, October, and November are the peak
months in the Lamigo Monkeys group because these are the months when the
World Baseball Classic (WBC), regular games, Taiwan Series, and Asian Baseball
Championship are held. Most of the fans from Lamigo Monkeys actively discuss
them on the Facebook group. However, December to February are trough months
when no competitions are held. Figure 9.2 demonstrates that March, May, and
August are the peak months in the Fubon Guardians group when fans focus on
the World Baseball Classic (WBC), all-star game voting, and regular games.

Table 9.8 Brand Attributes Over Time Periods (by Number)

Product-Related Non-Product-Related

Team/ Team/ Other Sport

Time Period Star Player Sport Game Non-Game People Issues

Lamigo Monkeys
Pre-season 16 6 6 0 0 6
Regular season 45 11 70 7 9 11
Post-season 37 9 55 6 13 9
Total 98 26 131 13 22 26
Fubon Guardians
Pre-season 134 25 79 11 27 25
Regular season 223 15 142 13 87 11
Post-season 155 22 156 9 42 20
Total 512 62 377 33 156 56
 Analyzing Sport Fan Facebook Pages 173

Figure 9.2 Percentage of Posts and Engagement by Teams.

Multicollinearity was measured by the variance inflation factor (VIF) and the
conditional index (CI). The findings showed that no multicollinearity occurred
because the highest value of VIF was 3.09 (less than 4) (Hair et al., 2010), and
the value of CI was 17.41 (less than 30) (Belsley, 1991). The model for post likes
is significant as a whole (F = 106.28, p < .01) and explains the variance of the
dependent variable reasonably well (R2 = 15.1%, adj. R2 = 14.9%). The pre-season
posts were significantly and negatively related to the number of likes (β pre − game =
−0.087, p < .01), whereas the regular-season posts were significantly and positively
related to the number of likes (β game = −0.051, p < .01). The low (β photo = 0.170,
p < .01) and high (β video = 0.123, p < .01) levels of vividness were significantly and
positively related to the number of likes but did not confirm hypothesis 1. The
high-level interactive post characteristics (i.e., “link”) were significantly related
174 S.-C. Ma et al.

to the number of likes (βlink = 0.397, p < .01), supporting hypothesis 2. The posts
talking about product-related attributes were significantly related to the number
of likes (β product = 0.031, p < .05), supporting hypothesis 3. However, the posts relat-
ing to non-product-related attributes were not significantly related to the number
of likes and thus did not support hypothesis 4 (Table 9.9).
The model for post comments was significant as a whole (F = 34.03, p < .01)
and explained the variance of the dependent variable reasonably well (R2 = 5.4%,
adj. R2 = 5.2%). The pre-season posts were not significantly related to the number
of comments, whereas the regular-season posts were significantly and positively
related to the number of comments (β game = 0.063, p < .01). The low (β photo =
−0.173, p < .01) and high (β video = −0.237, p < .01) levels of vividness are signifi-
cantly and negatively related to the number of comments, rejecting hypothesis 1.
The high-level interactive post characteristics (i.e., “link”) were significantly and

Table 9.9 Hierarchical Regression Analysis Results (DV: Number of Likes)

Model 1 Model 2 Model 3 Model 4

Variables β(t) β(t) β(t) β(t)

Team
Lamigo Monkeys — — — —
dummy (baseline)
Fubon Guardians 0.27 (19.18***) 0.27 (19.44***) 0.22 (16.39***) 0.23 (16.04***)
dummy
Time period
Post-season — — — —
dummy (baseline)
Pre-season dummy −0.07 (− 4.08***) −0.09 (−5.23***) −0.09 (−5.14***)
Regular-season 0.03 (1.41) 0.04 (2.50**) 0.05 (2.80***)
dummy
Post type
Text dummy — — — —
(baseline)
Photo dummy 0.17 (7.95***) 0.17 (7.80***)
Video dummy 0.13 (6.44***) 0.12 (6.37***)
Link dummy 0.40 (17.20***) 0.40 (16.92***)
Brand attributes
Product & — — — —
non-product
dummy (baseline)
Product dummy 0.03 (2.04**)
Non-product −0.02 (−1.23)
dummy
Constant 97.06 91.10 −1.61 −6.07
Adjusted R2 0.071 0.078 0.148 0.149
F-value 367.71*** 137.12*** 140.32*** 106.28***

** p < .05.
*** p < .01.
 Analyzing Sport Fan Facebook Pages 175

negatively related to the number of comments (βlink = −0.322, p < .01), rejecting
hypothesis 2. The posts associated with product-related attributes were signifi-
cantly related to the number of comments (β product = 0.062, p < .05), whereas the
posts associated with non-product-related attributes were not significantly related
to the number of comments, supporting hypothesis 3 and rejecting hypothesis 4
(Table 9.10).
The model for post shares was significant as a whole (F = 23.10, p < .01) and
explained the variance in the dependent variable reasonably well (R2 = 3.7%,
adj. R2 = 3.6%). The pre-season and regular-season posts were not significantly
related to the number of shares. The low (β photo = 0.053, p < .05) and high
(β video = 0.189, p < .01) levels of vividness were significantly and positively related
to the number of shares, supporting hypothesis 1. The high-level interactive
brand post characteristics (i.e., “link”) were significantly related to the number

Table 9.10 Hierarchical Regression Analysis Result (DV: Number of Comments)

Model 1 Model 2 Model 3 Model 4

Variables β(t) β(t) β(t) β(t)

Team
Lamigo Monkeys — — — —
dummy (baseline)
Fubon Guardians 0.07 (5.01***) 0.07 (5.11***) 0.10 (6.98***) 0.11 (7.19***)
dummy
Time period
Post-season — — — —
dummy (baseline)
Pre-season dummy −0.03 (−1.50) −0.02 (−1.36) −0.02 (−1.22)
regular-season 0.05 (2.85***) 0.04 (2.33**) 0.06 (3.28***)
dummy
Post type
Text dummy — — — —
(baseline)
Photo dummy −0.17 (−7.34***) −0.17 (−7.51***)
Video dummy −0.24 (−11.52***) −0.24 (−11.60***)
Link dummy −0.32 (−12.78***) −0.32 (−13.00***)
Brand attributes
Product & — — — —
non-product
dummy (baseline)
Product dummy 0.06 (3.90***)
Non-product −0.02 (−0.99)
dummy
Constant 13.06 4.12 49.02 42.28
Adjusted R2 0.005 0.010 0.049 0.052
F-value 25.12*** 16.67*** 42.00*** 34.03***

** p < .05.
*** p < .01.
176 S.-C. Ma et al.

Table 9.11 Hierarchical Regression Analysis Result (DV: Number of Shares)

Model 1 Model 2 Model 3 Model 4

Variables β(t) β(t) β(t) β(t)

Team
Lamigo Monkeys — — — —
dummy (baseline)
Fubon Guardians 0.09 (6.25***) 0.09 (5.28***) 0.06 (4.43***) 0.07 (4.50***)
dummy
Time period
Post-season dummy — — — —
(baseline)
Pre-season dummy −0.01 (−0.63) −0.01 (−0.71) −0.01 (−0.73)
regular-season dummy 0.01 (0.64) 0.02 (0.96) 0.02 (1.08)
Post type
Text dummy (baseline) — — — —
Photo dummy 0.05 (2.22**) 0.05 (2.29**)
Video dummy 0.19 (9.14***) 0.19 (9.17***)
Link dummy 0.19 (7.78***) 0.20 (7.85***)
Brand attributes
Product & non- — — — —
product dummy
(baseline)
Product dummy 0.002 (0.10)
Non-product dummy 0.02 (1.14)
Constant 0.50 0.46 −0.02 −0.06
Adjusted R2 0.008 0.008 0.036 0.036
F-value 39.08*** 13.71*** 30.58*** 23.10***

** p < .05.
*** p < .01.

of shares (βlink = 0.196, p < .01), in support of hypothesis 2. The post talking
about “product” and “non-product” was not significantly related to the number of
shares, rejecting hypotheses 3 and 4 (Table 9.11).

Discussion
The purpose of this study was to examine how different post characteristics and
brand attributes affect fans’ engagement (i.e., like, comment, and share) by season
periods on two fan-created Facebook pages for professional baseball teams. This
study extends to the sport management literature (Anagnostopoulos et al., 2018;
Parganas et al., 2015) by providing an opportunity to better understand how sport
fans interact with brand attributes and post characteristics on different social
media platforms in different professional sport (baseball) and culture (East Asia)
contexts. Based on the findings, professional teams can implement strategic plan-
ning for promoting team brand and products to sport fans via specific post types
on their social media.
 Analyzing Sport Fan Facebook Pages 177

The results show that both fan groups utilized links the most in the pre-
season, focusing on the topic of spring training. The findings are consistent with
previous studies (Anagnostopoulos et al., 2018; Clavio, 2013), which contended
that sport fans value ‘behind the scenes’ information about their team and play-
ers. This study offers team marketing managers additional information about
which post type (i.e., links) fans mostly adopted. During the regular season and
post-season, Lamigo Monkeys’ fans utilized photos the most and discussed teams/
games; Fubon Guardians’ fans utilized links the most and discussed star play-
ers beyond their favorite team, followed by teams/games and other people (e.g.,
coaches). Thus, this may suggest that Lamigo Monkeys’ fans pay more attention
to their favorite team (i.e., product-related attributes), implying that their fans
may have higher team identification (Byon et al., 2013). Fubon Guardians’ fans
focus on star players in CPBL and other professional baseball leagues (e.g., Major
League Baseball, Nippon Professional Baseball). It is reasonably speculated that
fans of the Fubon Guardians were attracted to star players of other teams because
the Fubon Guardians did not enter the post-season games. It is similar to Achen
et al.’s (2018) research that during the off-season period fans are still curious about
specific players’ life outside sports. We suggest that Fubon Guardian’s manager
could draw fans’ attention by posting attractive and interactive content, such as
behind-the-scenes stories (Achen et al., 2018; Li et al., 2019), to encourage fans’
ongoing engagement in particular when the post-season time is not available.
More importantly, in terms of post characteristics, it is suggested that photos
and links are the most favored ways for communicating and managing the team
brand, which sheds fresh light for practitioners and academia.
The results also revealed that, compared to the post-season period, the posts
published in the regular season can enhance the number of likes, whereas the
posts published in the pre-season period were found to be negatively related to the
number of likes. Overall, links (high level of interactivity) could result in a high
level of engagement (likes), followed by photos (low level of vividness) and videos
(high level of vividness). However, previous studies showed that there is no influ-
ence on links because to see them, it is necessary to go through another external
page (de Vries et al., 2012; Kwok & Yu, 2013; Liu et al., 2017; Osokin, 2019). This
might be explained by the fact that both groups of fans spent much time click-
ing through external websites relating to baseball news about CPBL games. As
noted by Cvijikj and Michahelles (2013), photos could attract the highest engage-
ment, followed by text, video, and links. Geurin and Burch (2017) also found
that photo posts mentioning the brand could evoke greater engagement since
fans are impassioned about the brand. Consequently, sport teams could consider
posting links or photos to promote the number of likes among fans. Videos also
positively encouraged fans to generate post likes, as indicated by previous stud-
ies (Osokin, 2019; Liu et al., 2017; de Vries et al., 2012). The findings also showed
that, compared to the combination brand attribute, product-related attributes (e.g.,
star player, team/game) alone can positively influence the number of likes, which
have been associated with higher purchase intentions and are a factor influencing
178 S.-C. Ma et al.

consumption behavior (Gladden & Funk, 2002; Kaynak et al., 2008). Besides, the
posts concentrated on product-related attributes could stimulate engagement by fol-
lowers (Geurin & Burch, 2017), and responses in terms of liking product-related
attributes can be higher than those of non-product-related attributes in the context
of Instagram for professional football (Anagnostopoulos et al., 2018). Thus, one
idea emerging from this research is that the sport teams can combine star players or
competitions with new product launches to create discussions.
The findings showed that, compared to the post-season period, the posts pub-
lished in the regular season can enhance the number of likes and comments.
Specifically, Lamigo Monkeys was eligible to play in the Taiwan Series, so fans
still paid attention to the team. However, Fubon Guardians did not enter the
post-season games, so their fans were easily distracted from their favorite team by
other hot issues (e.g., the Asia Winter Baseball League, Asia Professional Baseball
Championship). Li et al. (2019) indicated that the pre-season and in-season periods
are good times to attract new followers in emerging markets. As a result, it may be a
warning to those teams with poor league standings. We suggest that the manager of
Fubon Guardians should actively create topics that talk about the team and player
information to maintain discussion and fan support (Li et al., 2019).
The results of this study indicate that, compared to other characteristics (i.e.,
photos, videos, links), placing only a text can increase the number of comments,
which is practically consistent with de Vries et al. (2012) and Liu et al. (2017).
An explanation for this might be that the text post would let brand fans focus
on the topic and comment on the brand post directly (Cvijikj & Michahelles,
2013; Gyulai, 2016). Previous research (Moe & Trusov, 2011) also shows that dis-
criminating opinions in posted comments can bring about subsequent comments.
Followers may view them as a part of the community because they engage with
positive and negative arguments. The emotional response from fans not only
shows a fondness for their favorite teams but also displays team support (Chang,
2019). Significantly, the content analysis revealed that fans are most likely to
comment on product-related attributes such as games and star players, consistent
with previous research (Santos, 2019). Our findings suggest that, for professional
baseball teams to further understand fans, they should post with text.
Our findings demonstrated that post characteristics can enhance fan engage-
ment with the number of shares. Previous studies (Eagleman, 2013) indicated that
interactive posts can promote more fan interaction. This result is inconsistent
with previous research on brand posts in convenience stores (Liu et al., 2017),
which only discovered that call to act (e.g., collecting stickers or lucky draw) can
induce the number of shares. This is in line with Osokin’s (2019) perspective
that users prefer short and informative content. By knowing that posting links or
videos is one way to effectively increase the number of shares and fans’ comments
mostly related to star players and games, sport teams and their sponsors can have
their products and promotion activities highly exposed through links and videos
on social media. This is one of several ideas about using social media as a platform
for promoting teams and sponsors’ products to online fans.
 Analyzing Sport Fan Facebook Pages 179

Some limitations of this study should be acknowledged. First, Facebook pages

represent only one social media platform for examination, and therefore future
research should incorporate other platforms (e.g., Instagram, Twitter, etc.).
Second, every professional baseball group has different input costs, and therefore
future research should examine the relationship between costs and the success of
the team. Third, importantly, we were unable to perform demographic analysis
(i.e., age and gender) mainly because of the Personal Information Protection Act.
Future studies should seek other methods to fully understand participants’ demo-
graphic backgrounds. Fourth, the number of followers is a dynamic value; we
recommend that future studies could analyze the relationship between different
values. Fifth, in order to explore comprehensive effects, future studies could use
different statistical methods (e.g., supervised learning, logistic regression, random
forest, support vector regression, etc.). Lastly, additional factors such as the size of
the professional baseball fan groups, the frequency of posts, post length and post
time (i.e., workday and peak hours) differed. Future studies should consider this as
one of the sources of variance in the data analysis.

Acknowledgment
This study was financially supported by the Ministry of Science and Technology,
Taiwan (109PGFA0900691). We want to thank Dr. Yue Tan, who assisted in the
data collection and comments.

References
Abeza, G., O’Reilly, N., & Séguin, B. (2019). Social media in relationship marketing:
The perspective of professional sport managers in the MLB, NBA, NFL, and NHL.
Communication & Sport, 7(1), 80–109.
Abeza, G., O’Reilly, N., Seguin, B., & Nzindukiyimana, O. (2017). Social media as a
relationship marketing tool in professional sport: A netnographical exploration.
International Journal of Sport Communication, 10, 325–358.
Achen, R. M., Kaczorowski, J., Horsmann, T., & Ketzler, A. (2018). Exploring off-season
content and interaction on Facebook: a comparison of US professional sport leagues.
International Journal of Sport Communication, 11(3), 389–413.
Anagnostopoulos, C., Parganas, P., Chadwick, S., & Fenton, A. (2018). Branding in pic-
tures: Using Instagram as a brand management tool in professional team sport organisa-
tions. European Sport Management Quarterly, 18(4), 413–438.
Bauer, H. H., Stokburger-Sauer, N. E., & Exler, S. (2008). Brand image and fan loyalty
in professional team sport: A refined model and empirical assessment. Journal of Sport
Management, 22(2), 205–226.
Belsley, D. A. (1991). Conditioning diagnostics: Collinearity and weak data in regression. New
York, NY: John Wiley.
Byon, K. K., Zhang, J. J., & Baker, T. A. (2013). Impact of core and peripheral service
quality on consumption behavior of professional team sport spectators as mediated by
perceived value. European Sport Management Quarterly, 13(2), 232–263.
180 S.-C. Ma et al.

Chang, Y. (2019). Spectators’ emotional responses in tweets during the Super Bowl
50 game. Sport Management Review, 22(3), 348–362.
Cho, C. H. (1999). How advertising works on the WWW: Modified elaboration likeli-
hood model. Journal of Current Issues & Research in Advertising, 21(1), 34–50.
Clavio, G. (2013). Emerging social media and applications in sport. In P. Pedersen (Ed.),
Handbook of sport communication (pp. 259–268). New York, NY: Routledge.
Coyle, J. R., & Thorson, E. (2001). The effects of progressive levels of interactivity and
vividness in web marketing sites. Journal of Advertising, 30(3), 65–77.
CPBL (2020). About CPBL. Retrieved from http://www.cpbl.com.tw/footer/fans.html
Cvijikj, I. P., & Michahelles, F. (2013). Online engagement factors on Facebook brand
pages. Social Network Analysis and Mining, 3(4), 843–861.
De Vries, L., Gensler, S., & Leeflang, P. S. H. (2012). Popularity of brand posts on brand
fan pages: An investigation of the effects of social media marketing. Journal of Interactive
Marketing, 26(2), 83–91.
Eagleman, A. N. (2013). Acceptance, motivations, and usage of social media as a market-
ing communications tool amongst employees of sport national governing bodies. Sport
Management Review, 16(4), 488–497.
Fennis, B. M., & Stroebe, W. (2010). How consumers acquire and process information
from advertising. The Psychology of Advertising (p. 51). London, England: Psychology Press.
Filo, K., Lock, D., & Karg, A. (2015). Sport and social media research: A review. Sport
Management Review, 18(2), 166–181.
Geurin, A. N., & Burch, L. M. (2017). User-generated branding via social media: An
examination of six running brands. Sport Management Review, 20(3), 273–284.
Gladden, J. M., & Funk, D. C. (2001). Understanding brand loyalty in professional sport:
Examining the link between brand associations and brand loyalty. International Journal
of Sports Marketing and Sponsorship, 3(1), 54–81.
Gladden, J. M., & Funk, D. C. (2002). Developing an understanding of brand associations
in team sport: Empirical evidence from consumers of professional sport. Journal of Sport
Management, 16(1), 54–81.
Goldfarb, A., & Tucker, C. (2011). Online display advertising: Targeting and obtrusive-
ness. Marketing Science, 30(3) 389–404.
Gunawan, D. D., & Huarng, K. H. (2015). Viral effects of social network and media on
consumers’ purchase intention. Journal of Business Research, 68(11), 2237–2241.
Gyulai, D. (2016). A qualitative investigation of USA weightlifting Facebook follower
engagement during a world event, Master’s Theses, 4687.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis
(7th ed.). New Jersey: Pearson Prentice Hall.
Kaynak, E., Salmon, G., & Tatoglu, E. (2008). An integrative framework lining brand asso-
ciations and brand loyalty in professional sports. Brand Management, 15(5), 336–357.
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand
equity. Journal of Marketing, 57(1), 1–22.
Kwok, L., & Yu, B. (2013). Spreading social media messages on Facebook: An analysis of
restaurant business-to-consumer communications. Cornell Hospitality Quarterly, 54(1),
84–94.
Li, B., Dittmore, S. W., Scott, O. K., Lo, W. J., & Stokowski, S. (2019). Why we follow:
Examining motivational differences in following sport organizations on Twitter and
Weibo. Sport Management Review, 22(3), 335–347.
 Analyzing Sport Fan Facebook Pages 181

Liu, H.-W., Lin, Y.-L., & Wang, C.-H. (2017). Social media analytics for convenience
stores: The effects of media characteristics and marketing appeals on customer engage-
ment. Journal of Management and Business Research, 34(4), 633–657.
Liu, Y., & Shrum, L. J. (2002). What is interactivity and is it always such a good thing?
Implications of definition, person, and situation for the influence of interactivity on
advertising effectiveness. Journal of Advertising, 31(4), 53–64.
Luarn, P., Lin, Y. F., & Chiu, Y. P. (2015). Influence of Facebook brand-page posts on
online engagement. Online Information Review, 39(4), 505–519.
Malhotra, A., Malhotra, C. K., & See, A. (2013). How to create brand engagement on
Facebook. MIT Sloan Management Review, 54(2), 18–20.
Moe, W. W., & Trusov, M. (2011). The value of social dynamics in online product ratings
forums. Journal of Marketing Research, 48(3), 444–456.
O’Reilly, L. (2015). Posting a photo is the worst way to get people to see your
Facebook posts. Business Insider, 27. Retrieved from https://www.businessinsider.com/
facebook-photos-worst-for-organic-reach-socialbakers-video-text-2015-2
Osokin, N. (2019). User engagement and gratifications of NSO supporters on Facebook:
Evidence from European football. International Journal of Sports Marketing and Sponsorship,
20(1), 61–80.
Parganas, P., Anagnostopoulos, C., & Chadwick, S. (2015). ‘You’ll never tweet alone’:
Managing sports brands through social media. Journal of Brand Management, 22,
551–568.
Potter, W. J., & Levine-Donnerstein, D. (1999). Rethinking validity and reliability in
content analysis. Journal of Applied Communication Research, 27(3), 258–284.
Santos, T. O., Correia, A., Biscaia, R., & Pegoraro, A. (2019). Examining fan engagement
through social networking sites. International Journal of Sports Marketing and Sponsorship,
20(1), 163–183.
Statista. (2020a). Number of social network users worldwide from 2010 to 2023.
Retrieved from https://www.statista.com/statistics/278414/number-of-worldwide-social-
network-users/
Statista. (2020b). Most popular social networks worldwide as of April 2020, ranked by
number of active users. Retrieved from https://www.statista.com/statistics/272014/
global-social-networks-ranked-by-number-of-users/
Tafesse, W. (2015). Content strategies and audience response on Facebook brand pages.
Marketing Intelligence & Planning, 33(6), 927–943.
Watanabe, N., Yan, G., & Soebbing, B. P. (2015). Major League Baseball and Twitter
usage: The economics of social media use. Journal of Sport Management, 29(6), 619–632.
Watson, P. F., & Petrie, A. (2010). Method agreement analysis: A review of correct meth-
odology. Theriogenology, 73(9), 1167–1179.
Chapter 10

Marketing Obstacles Facing

Obstacle Racing
Jason Reese

Introduction
According to McKay and McKay (2020), obstacles have been used during physical
activities and for military purposes since ancient times. We have seen a transition
from the uses of obstacles by the ancient Greeks, Romans, and military around
the world to the novel physical challenge we see today in major sporting events
that combine exercise, entertainment, music, and all-day festivals (Fischer, 2015;
McKay & McKay, 2020; Mullin, 2012). The popularity of these events can be
traced to various theories. Weedon (2015) suggests that part of the appeal for
obstacle racing is the “camaraderie” shared by participants. Mullin (2012) sug-
gests that “two of the most prominent theories of physical activity motivation,
achievement goal theory and self-determination theory (SDT), may at least partly
explain their appeal” (p. 103).
The “key players in the obstacle course business” are Spartan Race, Warrior
Dash, Tough Mudder, and Rugged Maniac (Fischer, 2015, p. 17). Each of these
race promoters has entry fees ranging from $50 to upward of $200. Spartan Race
puts on the most races at over 150 a year. Each of these promoters boasts hun-
dreds of thousands of participants each year (Fischer, 2015). The growth of the
obstacle racing industry is large. By 2015, about six years after the entrance into
the market, the number of runners reached nearly five million all time (Fischer,
2015). Early on, the growth from year to year was spectacular. From 2010 to 2011,
the growth rate sextupled. In 2012, the growth rate within the industry began
to slow. In that year, the growth rate doubled. The growth rate was only 48% in
2013 and even less in 2014, 24% (Fischer, 2015). In 2009, the estimated revenue
from obstacle racing was $15.9 million. Estimated revenue increased in 2014 to
$362 million (Fischer, 2015). This has led to the development of the World OCR
in 2014, the sport’s governing body, which “had 89 national federations among its
membership” (Radnedge, 2019, p. 1). In addition, the increased prize money has
led to more athletes participating. For example, in 2019, those participating in the
three-part obstacle racing event of the Ultra World Championship, the Trifecta
World Championship, and the Spartan World Championship can win upward of
$1 million (Radnedge, 2019). Because of the increase in participant rates, and the

DOI: 10.4324/9781003302476-10
 Marketing Obstacles Facing Obstacle Racing 183

increase in the number of obstacle racing promoters, it is important to understand

consumer’s perceptions. These perceptions can help promoters better understand
their reason for purchase and what they are looking for in obstacle races.
One of the major concerns for promoters of obstacle races is the recent decline
in participation. For example, many executives of the larger promoters are openly
discussing the maturity of the obstacle racing industry only six years after its
entrance into the market (Fischer, 2015). This concern, coupled with the fact
that approximately 40% of the runners of major obstacle races are repeat custom-
ers (Fischer, 2015), marketers of these types of events should be aware of consumer
perceptions, to better ensure repurchasing. The purpose of this study is two-fold:
a) to examine a case of an annual participant-focused obstacle sporting event and
better understand the cause of declined participation and b) determine the most
appropriate pricing strategies for participant sporting events like these.

Review of Literature

Price Fairness
The associations and emotions customers have when they evaluate the price they
paid is the foundation of price fairness (Xia et., 2004). In addition, price fairness is
arguably a consequence of price comparisons with consumers’ internal reference
prices (Petrick, 2005) and other providers’ prices (Xia et al., 2004). For exam-
ple, Malc et al. (2016) “confirms that social comparisons play an integral part in
determining fairness” (p. 3693). Price fairness is defined as “a consumer’s assess-
ment and associated emotions of whether the difference (or lack of difference)
between a seller’s price and the price of a comparative other party is reasonable,
acceptable, or justifiable” (Xia et al., 2004, p. 3). Firms should be concerned price
fairness as it can have a great influence on future purchase behavior. For example,
“perceptions of price unfairness may lead to negative consequences for the seller,
including buyers leaving the exchange relationship, spreading negative informa-
tion, or engaging in other behaviors that damage the seller” (Xia et al., 2004, p. 1).
Greenwell et al. (2008) argued that it is often more difficult to study price
fairness in the context of the sport due to the strong emotional connection con-
sumers in the sport industry often have with organizations. In addition, they also
suggested that due to fewer consumer options, studies investigating price fair-
ness in a sport context are unique. Regardless of the reason for its investigation,
sport industry scholars and practitioners should be aware of the importance and
uniqueness of price fairness perceptions in the sport industry. Price fairness is
investigated through the lens of four theories: equity theory, dual entitlement
theory, distributive justice, and procedural justice (Xia et al., 2004). Equity theory
is the core of all these theories. This theory suggests that there should be an equi-
table disbursement of all resources to appropriate exchange parties, not necessar-
ily equally (Kim et al., 2009). In other words, what the consumer gives up in the
exchange should be equitable to the benefit they receive from the producer/firm.
184 J. Reese

Distributive justice advances equity theory by comparing the outcome of the

exchange without taking into account the existing relationship the exchange
parties have with each other (Oh, 2003; Park et al., 2010; Xia et al., 2004). Here,
it takes into account the relationship and how it could change the perception
of equity in the exchange (Kim et al., 2009). This is important to the discus-
sion in the current study, as the knowledge and relationship the runners have
with the non-profit promoter may change their perceptions of the event. This is
reflected later, as hierarchical regression analysis is used in the statistical analyses,
controlling for knowledge of the non-profit promoter. Dual entitlement theory
investigates how the customer’s price is aligned with the firm’s expenses (Bolton
& Alba, 2006). In short, the theory suggests that all exchange parties are entitled
to an equitable outcome. Haws and Bearden (2006) suggest the theory proposes a
fair (reasonable) profit for the firm and a fair (reasonable) price for the consumer.
In other words, price increases should be reflecting increases in costs (Bolton &
Alba, 2006).

Perceived Value
Coming from the marketing and consumer behavior literature (e.g., Han &
Kwon, 2009; Helkkula & Kelleher, 2010; Itani et al., 2019; Kwon et al., 2007;
Parasuraman, 1997; Parasuraman & Grewal, 2000; Zeithaml, 1988), perceived
value is the comparison of the product received and what is given by the con-
sumer. Zeithaml’s (1988) seminal definition of perceived value is “the customer’s
overall assessment of the utility of a product based on perceptions of what is
received and what is given” (p. 14). Arguably, the core variable in examining
perceived value is the perception of price and quality (Petrick, 2002). Perceived
value is often considered the core of consumer decisions making. For example,
some argue that “perceived value is the most important indicator of future pur-
chase behavior for services” (Petrick, 2004, p. 29). Furthermore, other scholars
argue that having a good understanding of perceived value can lead to a competi-
tive edge (Parasuraman, 1997). Itani et al. (2019) suggest that perceived value has
both indirect and direct positive impact on customer engagement. Investigating
perceived value can be approached in several ways (Zeithaml, 1988). In the con-
text of the current study, examining perceived value as a comparison of what the
consumer receives and what they give up is most appropriate. If the product is not
providing an adequate return on the consumer’s investment in time/money, they
will likely perceive the experience as having little value.

Satisfaction
In today’s economy, sport organizations are constantly looking for ways to meet
the needs and wants of consumers. According to Greenwell (2007), attendance at
sporting events is commonly a problem for sport organizations. Firms understand
that customer satisfaction is commonly associated with word-of-mouth referrals
 Marketing Obstacles Facing Obstacle Racing 185

from current customers (Xu et al., 2006). In addition to word-of-mouth, customer

satisfaction has also been shown to be an indicator of a firm’s profit (Anderson &
Sullivan, 1993; Homburg et al., 2005). Therefore, firms have sought to understand
ways to improve and control customer satisfaction.

Customer Satisfaction Defined

The seminal pieces of literature on customer satisfaction and loyalty come from
Richard Oliver. Oliver (1999) suggested that customer satisfaction can be defined
as a pleasurable level of consumption-related fulfillment. In other words, “the con-
sumer senses that consumption fulfills some need, desire, goal, or so forth and that
this fulfillment is pleasurable” (p. 34). Customer satisfaction has also been seen
as “a comparison of expectations against a process or outcome” (Greenwell, 2007,
p. 7). Suh and Yi (2006) also suggest that the experience consumers have with a
good or service is compared with expectations, and satisfaction perceptions are
formed based on these expectations. This variation from the customer’s expecta-
tion is considered a disconfirmation of expectations (Anderson & Sullivan, 1993).

Transaction-Specific Satisfaction
Beccarini and Ferrand (2006) suggest that satisfaction can be evaluated by look-
ing at the relationship between a customer and firm during a specific or several
transactions. Some researchers have argued that customer satisfaction is a sum-
mation of the attitudinal judgments of a series of purchases (Xu et al., 2006).
However, some researchers have argued that customer satisfaction can be exam-
ined immediately following a single transaction or event (e.g., Olsen & Johnson,
2003). This type of customer satisfaction evaluation is termed transaction-
specific satisfaction. The alternative option would be to use a cumulative satisfac-
tion approach (Homburg et al., 2005). In the cumulative approach, the customer
is asked to evaluate their overall experiences with a good or service. The current
study is attempting to determine the overall evaluation of an obstacle event, and
therefore, the cumulative approach will be used.

Experiential Goods/Services and Customer Satisfaction

Sport goods and services are often experiential in nature. When you participate
in a sporting event, you are consuming an intangible good. In addition, sport
goods and services often bring with them emotional attachments that are some-
times viewed by society as being relatively irrational but acceptable. Therefore,
emotions should often be considered when evaluating satisfaction. Caro and
Garcia (2007) echo this idea when they suggest “emotional reactions associated
with the consumption experience are important in the determination of satisfac-
tion” (p. 71). This idea of emotional inclusion into the assessment of satisfaction
is not an isolated thought. Kwak et al. (2011) suggested that satisfaction is an
186 J. Reese

“attitude-like evaluative judgment” (p. 6). This would support Caro and Garcia’s
(2007) statement that it is especially important to include emotions when dealing
with services, specifically those experiential in nature (i.e., sport).

Customer Satisfaction & Repurchase Intentions

The relationship between satisfaction and repurchase intentions is one of the
most researched areas in the satisfaction literature. Oliver (1980) suggested that
customer satisfaction has traditionally been influenced by purchase decisions.
However, it really was not until the early 1990s when that researchers consistently
empirically investigated this relationship. In the sport marketing literature, several
researchers have examined and/or discussed this relationship. Tsuji et al. (2007)
investigated this link in their study of action sports, service quality, and satisfac-
tion. According to Tsuji et al. (2007), “peripheral service quality and satisfaction
were found to be significantly related to future intentions” (p. 199). Furthermore,
Trail et al. (2005) found that satisfaction influenced consumers’ behavioral inten-
tions. Finally, Murray and Howat (2002) posit that “satisfaction appears to be a
dominant antecedent of the future intentions of customers” (p. 37).

Purchase Intentions
Without re-patronage, firms will fail (Wilkinson, 2009). While this is likely
not considered a novel concept, it is important to understand the consumers’
intentions as it will provide marketers with an idea of how to create successful
campaigns. Using (re)purchase intentions as the dependent variable is common
among consumer perception studies. It provides the researcher the ability to better
understand and predict future behaviors. The theory of planned behavior (Ajzen,
1991), derived from the theory of reasoned action, is often considered the foun-
dation of purchase intention evaluations. The stronger the intention, the more
likely the consumer will behave in that manner. “A central factor in the theory
of planned behavior is the individual’s intention to perform a given behavior”
(p. 181). According to Tsuji et al. (2007), “measuring future intentions of attendees
is important for the continued success of an event” (p. 201). This is partly due to
the connection between repurchase intentions and perceived value. For example,
perceived value has been shown to influence golfer repurchase intentions (Petrick
et al., 1999). Scholars in other fields have also found the same perceived value/
purchase intention relationships.

Price Sensitivity
Understanding a consumer’s perception of price is important for firms attempting
to examine their future behavioral intentions and retention (Dean, Morgan, Tan,
2002). It is suggested that price sensitivity “is considered a component of behavio-
ral intentions” (Dean et al., 2002, p. 97). Petrick (2004) suggests, “Price sensitivity,
 Marketing Obstacles Facing Obstacle Racing 187

which is similar to price elasticity, is a measure of how much of an affect increases

in price will have on customer’s willingness to pay” (p. 465). In other words, price
sensitivity is examining how consumers respond to price changes. For example, a
price-sensitive consumer is one who purchases based on the price of the good or
service being offered, more likely needing a discount than a less price-sensitive
consumer (Petrick, 2005). Consumers who are price sensitive have been described
as “blue-collar fans tend(ing) to have less disposable income” (Rishe & Mondello,
2003, p. 73). Marketers have also used price sensitivity as a way to segment the
market, as “price perceptions often differ significantly across market segments”
(Bolton & Myers, 2003, p. 108). Even though these types of consumers base many
of their purchases decisions on price, when prices are equal among competitors
in a price-sensitive market, consumers will focus more on the product’s qualities
(Ainscough, 2006).

Research Questions
Based on the aforementioned literature and the current participation problem
facing the event being studied, the following research questions were developed:

1 Is the reason for the decline in participation across the three years of study
due to problems with event elements and consumer perceptions?
2 Does the price tier system currently being used for registration appropriate
for the event promoters, considering consumers’ perceptions of price (fairness
and sensitivity)?
3 What are the best predictors of future purchase behavior? In other words,
what measures can we used to better determine repurchase intentions?

Method
This study investigates consumer perceptions and behaviors related to an annual
fundraising obstacle racing event in the south region of the United States. The
event is organized and promoted by a junior chamber organization. The purpose
of the organization is to help their members develop leadership skills, as well as
give back to the local community through projects and donations to other local
non-profit organizations. The idea for this event was presented to the board of
directors by a few of their members. After planning the event for a year, the event
was held for the first time in 2013. The first event hosted 617 runners. In 2014, the
event grew to 834 runners. It peaked in 2015 with 1,050 runners. Since that year,
the participation has declined to 885 in 2016 and 627 in 2017.
The literature surrounding sporting event demand suggests that some uncon-
trollable variables have an impact on demand for sporting events (Rascher et al.,
2007). In the case of this event, each year, the weather has been pleasant, with
no rain during the event. Other elements out of the control of the event promot-
ers include community-wide activities that compete for the runners’ interests, as
188 J. Reese

well as volunteers’ availability. In 2017, a large community service event around

the local university occurred on the same day. This likely impacted the number
of volunteers, as well as some runners. Other than 2017, very few local external
elements impacted participation.

Price Tiers in Sport

The segmentation of goods and services offered by a firm is not uncommon in any
industry. Most often, these segments are formed by categorizing products by their
attributes and functions. The use of price tiers is often the way one can identify
categories. According to Sivakumar (2000), the segmentation of price tiers can
be based on retailer strategy, size, competition of product attributes, store type,
consumer characteristics, manufacturer strategy, or environmental factors. Some
scholars argue that the segmentation of price tiers is based on quality and/or price
(Rao, 2005; Sivakumar, 2000)
Due to the dynamic sport industry, the types of products offered vary greatly.
Price tiers could be based on the location of a seat in a facility. The tiers could be
based on access to a fitness facility during certain times. For events like the one
being investigated in this study, price tiers can be set up based on the registration
date. Regardless of the product being offered, price tiers are defined as “the variety
of prices that are typically offered at any given time” (Moe & Fader, 2009, p. 74).
Moe and Fader (2009) also suggest, “failure to account for price tiers can lead to
exaggerated inferences about the role of price over time” (p. 73).

Participants
The sample in Study 1 consisted of participants of an annual obstacle racing
event across a three-year period, from 2015 to 2017. The data from Study 2 (2018)
race year are discussed later in the discussion section. The age of race participants
ranged from under 15 years old (approximately 16%) to over 50 years old (approxi-
mately 4%) during this three-year period. While the majority of runners (approxi-
mately 74%) during this time period lived within 50 miles of the race venue, some
runners traveled over 200 miles (one-way) to attend these events. The majority
of the runners for this event during this three-year period were female, approxi-
mately 57% of the race field.
The sample consisted of 143 runners in 2015, 87 runners in 2016, and 96 run-
ners in 2017. The combined three-year sample consisted of 184 females (56.3%)
and 75 males (23%) participants (note: 67 participants chose not to respond to
this question). The majority of the study participants were White, non-Hispanic
(N = 217, 66.6%). The study participants’ mean age was 35.36 years old (SD =
10.30). Approximately half (49.6%) of the sample had at least a college degree.
A little less than half of the sample (45.7%) had an annual household income of
over $50,000.
 Marketing Obstacles Facing Obstacle Racing 189

Research Design and Procedures

This longitudinal study began in 2015. Beginning at that time, the data were
collected from participants using an online survey-linked questionnaire. Within
a few days of each event, the event promoter emailed all race participants. In the
email, they thanked them for their participation in the event and asked that they
click the link embedded in the email and complete the questionnaire. After the
runner clicked the link, they were taken to an online questionnaire. The first
screen asked the runner to consent to participation. Each year, the questionnaire
consisted of questions related to several variables. These variables were suggested
by the research team, and the event promoter approved them prior to distribu-
tion. In addition, in the years 2016 and 2017, some questions were added per
request from the event promoters. These additional questions primarily dealt with
changes to the obstacles and/or event activities (e.g., medals, finish-line activities,
etc.). At the end of each questionnaire, the study participant was asked to provide
demographic information (e.g., age, gender, ethnicity, income, and education).
Finally, the event promoters offered a free entry fee to one participant at the end
of the study each year. If the study participant wanted to be considered for this,
they were asked to provide their email address at the end. They were allowed
to opt out of this part of the study. The research team then provided the list of
emails to the event promoter without being linked to the raw data at the end of
the study period each year.

Measures
The measures used in this study were developed based on extant literature, except
satisfaction questions related to race obstacles and knowledge of the non-profit
promoter. For example, study participants were asked to rate each race obstacle
(25–28 per year). All measures in this study were based on a 7-point scale, with 1
being “very dissatisfied” to 7 being “very satisfied”. In addition, a simple agree-
disagree scale for the runners’ knowledge and support for the non-profit and their
events was presented. The questions related to knowledge of the reason for the
organization, self-identification as a supporter, and intent to attend other events
hosted by the non-profit.
The measures for satisfaction were adapted from Greenwell et al. (2007), with
added event satisfaction elements from Tsuji et al. (2007). The price fairness meas-
ures were adapted from Xia et al. (2004). The measures for price sensitivity were
adapted from Petrick (2005). The perceived value measures were adapted from
Petrick (2002). Finally, the repurchase intentions were adapted from Hermann,
Xia, Monroe, and Huber (2007). According to the Cronbach’s alpha tests, no
issues with internal consistency were found (satisfaction = .840, price fairness =
.911, repurchase intentions = .853, price sensitivity = .830, knowledge of the non-
profit = .825, and perceived value = .913). In addition, to test for validity, Pearson
Product–Moment Correlations were used. As with internal consistency, there
190 J. Reese

were no issues with validity, as all relationships were moderate to strong positive
and significant (p < .001) for each measure.
To further test for validity and reliability, a factor analysis was conducted
on each construct. No issues were found in the analysis. For each analysis, the
Kaiser–Meyer–Olkin Measure of Sampling Adequacy (KMO) and Bartlett’s Test
of Sphericity revealed a factor analysis was appropriate (satisfaction KMO = .781
and Bartlett’s = .000, price fairness KMO = .677 and Bartlett’s = .000, repurchase
intentions KMO = .695 and Bartlett’s = .000, price sensitivity KMO = .653 and
Bartlett’s = .000, knowledge of the non-profit KMO = .738 and Bartlett’s = .000,
and perceived value KMO = .630 and Bartlett’s = .000). In addition, all values
within the correlation matrix were above .300 (satisfaction = all above .457, price
fairness = all above .635, repurchase intentions = all above .605, price sensitivity =
all above .481, knowledge of the non-profit = all above .727, perceived value = all
above .474), with none showing similarity (those close to 1.000).

Statistical Analyses
Four statistical analyses were used during this study. First, to examine demographic
information, participation, and mean scores, basic descriptive analysis was used.
Second, to better determine the reason for the decline in participation, an exami-
nation of multiple variables between years was employed. Here, a multivariate anal-
ysis of variance was conducted. Third, to analyze price sensitivity and price fairness
perceptions based on the time of registration, two separate analysis of variance
were conducted. Finally, to better predict a runner’s intent to return in future years
(repurchase intentions), while simultaneously controlling for a runner’s connection
with the non-profit, a hierarchical regression analysis was conducted.

Results
To help answer Research Question 1 (reason for the decline in participation), a
multivariate analysis of variance was conducted. To better determine the reason
for the decline in participation, a number of variables were compared between
years. The dependent variables were added based on the extant literature indi-
cating potential impact on future purchase behavior. These dependent variables
included both the event and obstacle satisfaction (Murray & Howat 2002; Tsuji,
et al., 2007; Trail et al., 2005), price fairness (Xia et al., 2004), perceived value
(Petrick, 2004), and knowledge of the non-profit promoter (Kim et al., 2009).
Here, multivariate analysis of variance (MANOVA) was used. The analysis
revealed there was indeed a significant multivariate effect, Wilks’ λ = .925, F (10,
500) = 1.999, p = .032. Due to this significant effect, the univariate statistics were
analyzed. Results are shown in Table 10.1. The results indicated that all depend-
ent variables were not significantly different from year and year, except price fair-
ness. To analyze the difference in price fairness perceptions between years, the
Duncan post hoc test was used. Results reveal the mean score for the 2017 group
 Marketing Obstacles Facing Obstacle Racing 191

Table 10.1 MANOVA for Event Perceptions

Dependent Variable Df F p Partial η2

Event satisfaction 2 .758 .345 .008

Obstacle satisfaction 2 .348 .504 .005
Price fairness 2 2.892 .036 .026
Perceived value 2 .368 .677 .003
Knowledge of the 2 .160 .923 .001
promoter
Error 254

(M = 5.78) was significantly lower than 2016 (M = 6.12) and 2015 (M = 6.10), but
the 2016 and 2015 score did not significantly differ from each other.
To help answer Research Question 2 (price tier effectiveness), an analysis
of variance was conducted. The four price tiers for each year were as follows:
(a) Tier 1 – $30 (December 15–February 1), (b) Tier 2 – $40 (February 2–March 1),
(c) Tier 3 – $50 (March 2–April 1), and (d) Tier 4 – $60 (April 2 – Race Day).
Results show that the effects of the registration date on price sensitivity is not
significant, F (3, 254) = .355, p = .785, partial η2 = .004. The mean scores for
runners in the four registration periods did not differ significantly from each
other. The mean scores ranged from 4.60 to 4.83 (with 7 being very price sen-
sitive). An analysis of variance was conducted. Results show that the effects
of the registration date on perceived price fairness is significant, F (3, 257) =
3.125, p = .026, partial η2 = .035. Results indicate that the mean values for
those who registered before the first price increase (M = 6.25) are significantly
greater than the values for those who registered in the final two prices’ higher
tiers (both M = 5.85). However, those who registered before the second price
increase (M = 6.17) did not differ significantly from the any other group.
To help answer Research Question 3, a hierarchical regression analysis was
conducted (Table 10.2). This event is held annually in a small rural community
of approximately 34,000 citizens in the south region of the United States. The
event promoter is a well-known philanthropic group that garners much respect
in the community. Therefore, some people in the community likely register for

Table 10.2 Hierarchical Regression Analysis

Model 1 Model 2

Variable B SE B β B SE B β

Knowledge of the promoter .234 .047 .341** .085 .033 .124*

Satisfaction .364 .063 .306**
Price fairness .066 .073 .066
Perceived value .482 .071 .493**

* p < .05.
** p < .001.
192 J. Reese

Table 10.3 MANOVA for Event Perceptions

Dependent Variable Df F p Partial η2

Event satisfaction 3 .758 .076 .020

Obstacle satisfaction 3 .348 .052 .022
Price fairness 3 2.892 .030 .026
Perceived value 3 .368 .842 .002
Knowledge of the 3 .160 .417 .008
promoter
Error 341

their events to support the work they do in the community. To better determine
the true effect satisfaction, perceived value, and price fairness perceptions have
on repurchase intentions, the knowledge of the non-profit promoter was included
as a control variable. Before an analysis of all three years of data could be run,
it was important to ensure the data between the years were similar. Therefore,
three separate hierarchical regressions were run, one for each year of data. The
2015 and 2017 hierarchical regression were both significant. However, the 2016
data were not significant at the .05 level (p = .066). Therefore, the 2016 data were
excluded from the multi-year hierarchical regression. The tests for multicollinear-
ity issues were performed and indicated there was a low level of multicollinearity
present (VIF = 1.118 for knowledge of the promoter, 1.405 for satisfaction, 2.587 for
price fairness, and 2.643 for perceived value). The control accounted for 12% of
the variance (p < .001) in repurchase intentions. After accounting for this effect,
the independent variables (satisfaction & perceived value) accounted for an addi-
tional 51% unique variance (p < .001). Both satisfaction (β = .306, p < .001) and
perceived value (β = .493, p < .001) were significantly associated with repurchase
intentions. However, perceived price fairness was not a significant predictor of
repurchase intentions (β = .066, p = .365) (Tables 10.3 and 10.4).

Discussion
To better determine the reason for the decline in participation, a number of varia-
bles were compared between years. All of the variables (overall event satisfaction,
obstacle satisfaction, perceived value, and knowledge of the non-profit promoter)

Table 10.4 Duncan’s Post Hoc Test

Subset

Year N 1 2

2017 72 5.776
2018 88 5.799
2015 115 6.101
2016 70 6.121
 Marketing Obstacles Facing Obstacle Racing 193

were not significantly different between years, except price fairness. In 2017, price
fairness perceptions were significantly lower than the previous two years. Due to
the consistent year-to-year drop in participation and the understanding of equity
theory, one would likely surmise that one of the causes in participation could
be unfair pricing tactics. However, the prices for the current event did not
change from across the three years studied, nor were the price tiers. In fact, med-
als were added to the 2017 race, adding an element that could potentially increase
price fairness perceptions.
To better understand why this variable showed this significance, the raw data
were reviewed. When looking at those participants that filled out the question-
naire, 24.7% of all respondents registered for the race during the highest-priced
period. However, among the respondents to the 2015 and 2016 questionnaires,
only 11.9% and 11.4% registered during the highest-priced period, respectively. As
the results of this study’s ANOVA revealed, the later the participant registered for
the event, the lower their price fairness perceptions went. Therefore, lower price
fairness perceptions could be due to the sample in 2017, rather than an overall
price fairness perception issue for the race over the studied three-year period.
Regardless of the reason for the price fairness significance in the MANOVA,
the results of the study are likely most interesting because of the insignificance of
the rest of the dependent variables. Extant literature suggests that consumer per-
ceptions of satisfaction (Murray & Howat, 2002; Oliver, 1980; Tsuji et al., 2007;
Trail et al., 2005), price fairness (Xia et al., 2004), and perceived value (Petrick,
2002) could all lead to changes in future participation. However, the results from
this study suggest that may not be true for this event promoter. In other words,
year after year, this event promoter had positive evaluations, yet the attendance
continued to decline. The results of the study suggest the decline is likely not
due to anything the event promoter is doing wrong, but more than likely, it is an
indictment on the obstacle racing industry.
This promoter used a price tier system based on the registration date. As the
date of the event gets closer, the price to register for the event increases. This
was done for a couple of reasons. First, the event promoter may be attempting to
attract consumers who are price sensitive (e.g., looking for the best deal). Second,
event promoters try to “lock-in” participants to better plan the event. When
grouped based on registration date, the price fairness perceptions decrease as the
date of the event gets closer. Those who registered before the first price increase
(M = 6.25) are significantly greater than the values for those who registered in the
final two higher-priced tiers (both M = 5.85). When grouped based on registration
date, the price sensitivity of runners is not significantly different between groups.
The mean scores ranged from 4.60 to 4.83 (with 7 being very price sensitive).
This would disagree with the event promoter’s idea that they are attracting
price-sensitive consumers early in the registration process.
If you couple the two findings above (lower price fairness perceptions as the
event gets closer and no significances in price sensitivity between registration
groups), one could argue that the price tiers for this event promoter are not
194 J. Reese

needed. In other words, if the price fairness is decreasing as the price goes up, but
the price sensitivity is constant, why have to price tiers at all? Why not set a fixed
price for the entire registration process? Two issues may come to bear for the event
promoter if they were to make this change: (1) the runners will just wait to the
last minute to register, and (2) the price fairness will be lower for every runner,
instead of just those who register later. These two issues are addressed below.
The first concern about removing the price tiers is that runners may just wait
to the last minute to register, leaving the promoter little time to plan for the size
of the race field. In the current situation, the facility being used by the promoter
does not change based on the number of runners in the race field. Individual heats
are capped at a predetermined number of runners. The number of volunteers will
not change based on the size of the race field. In addition, the race promoter plans
on the number of shirts and medals based on the number of registered runners as
the event comes closer. By setting a hard registration date for those that want to
ensure they receive a shirt, the promoter could still plan appropriately. In addi-
tion, the cost would be the same regardless of the registration date, allowing the
promoter to save the cost of T-shirts and turning that toward their profit. Finally,
comparing the number of registrations this event promoter had throughout its
existence, approximately 94% of participants registered before the “shirt guaran-
tee” deadline. The second concern about removing the price tiers is that all run-
ners will have lower price fairness perceptions, instead of just those who register
later. To examine this relationship in the current study, a hierarchical regression
analysis was used. When controlling for runners’ knowledge of the non-profit
promoter, satisfaction with the event and perceived value significantly predicted
repurchase intentions. However, in this same analysis, price fairness did not sig-
nificantly predict repurchase intentions. In other words, the perceived fairness of
the registration fee did not impact whether they planned to run in the race in
the future. In addition, even the group with the lowest price fairness perceptions
(those purchasing at the highest prices) had relatively high price fairness ratings
(M = 5.85, with 7 being the highest price fairness perception).
Following the results from the 2015–2017 data, price tiers were eliminated from
the pricing structure of the 2018 event. Instead, a flat rate of $50 for pre-race day
and $60 for race day registration were used. These prices were determined based
on the last price tier in 2013–2017. Below is a comparison of the 2013–2017 and
the 2018 price tiers. From 2013–2017, the price tiers were set up as follows: (a)
Tier 1 – $30 (December 15–February 1), (b) Tier 2 – $40 (February 2–March 1),
(c) Tier 3 – $50 (March 2–April 1), and (d) Tier 4 – $60 (April 2 – Race Day). In
2018, the pricing structure was $50 with discounts (December 15–April 1) and
$50 with a $10 late registration fee (April 2 – Race Day). In addition to the price
changes, an increase in marketing spending for 2018 occurred. In 2017, $1,740.64
was spent on advertising, with almost all being spent on targeted social media
spending. The spending increased 87.97% in 2018 to a total of $4,419.49. The
result was a 162-runner increase (25.84%) for a total of 789 runners. The reve-
nue from increased registration fees was $6,691.25 (33.63%). The price per runner
 Marketing Obstacles Facing Obstacle Racing 195

increased from $26.40 in 2017 to $29.46 in 2018. As there was a change in price
methodology in 2018, an additional round of data collection and analysis were
conducted (Study 2). The methodology was identical to Study 1. The Study 2
(2018) sample consisted of 144 runners, 74.7% females and 16% males, with
36.8% choosing not to answer. The majority of Study 2 (2018) study participants
were White, non-Hispanic (N = 78, 85.7%). The study participants’ mean age
was 37.13 years old. Over half (63.4%) of the sample had at least a college degree
and (62.2%) had an annual household income of over $50,000.
In Study 2, a MANOVA was conducted to examine the differences in consumer
perceptions, including the 2018 questionnaire data. The dependent variables
were the same as the previous MANOVA in this study and included overall event
satisfaction, obstacle satisfaction, price fairness, perceived value, and knowledge
of the non-profit promoter. The analysis revealed there was indeed a significant
multivariate effect, Wilks’ λ = .896, F (15, 930) = 2.517, p = .001. Due to this signif-
icant effect, the univariate statistics were analyzed. The results indicated that all
dependent variables were not significantly different from year to year, except price
fairness (p =.030). To analyze the difference in price fairness perceptions between
years, the Duncan post hoc test was used. Results are displayed in Table 10.4. The
results reveal the mean score for the 2017 (M = 5.78) and 2018 (M = 5.80) groups
were significantly lower than 2016 (M = 6.12) and 2015 (M = 6.10), but the 2015
and 2016 subsets were not significantly differing from each other. The results
from Study 2 (2018) data analysis might suggest that a change from price tiers
to a fixed price with price promotions did not change price fairness perceptions,
nor did it decrease the price paid per runner. In fact, it could be argued that the
price per runner increase was due to price changes, however, this statement is not
controlling for the increase in marketing expenditures.
There are a few study limitations. First, the study is contextually limited.
While the data were collected longitudinally, they were only collected at one
event. This event was unique in that the event promoters are a non-profit vol-
unteer organization. Their purpose for hosting the event was fundraising for
the community. Therefore, their motives are likely different from for-profit race
promoters. Another limitation to this study is the exclusion of the 2016 data
from the hierarchical regression analysis. Having a complete three-year period
would give more strength to the results. In addition, while the same variables
were measured from year to year, changes in the event were not controlled for
in the analyses. For example, each year, the event promoters added or removed
obstacles, added medals in 2017, and changed their advertising approach
from year to year. These elements could impact the consistency of responses
between years. Promotions each year typically followed changes in the price
(tier). However, because there is no precise information available regarding the
timing of promotions each year, it may be a combination of promotion timing
and price tiers that made the difference between years. This is an additional
limitation. Future research could focus on a larger set of obstacle racing events.
Comparisons across multiple regions, multiple event promoters (both for- and
196 J. Reese

non-profit), as well as types of non-traditional running events would be benefi-

cial. Understanding consumer perceptions and behaviors is important for mar-
keters of these types of events.

References
Ainscough, T. L. (2006). The effect of brand, agent, and price on consumer evaluation of
travel services. Journal of Travel & Tourism Marketing, 19, 39–48.
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50, 179–211.
Anderson, E. W. & Sullivan, M. W. (1993). The antecedents and consequences of cus-
tomer satisfaction for firms. Marketing Science, 12(2), 125–143.
Beccarini, C., & Ferrand, A. (2006). Factors affecting soccer club season ticket holders’
satisfaction: The influence of club image and fans’ motives. European Sport Management
Quarterly, 6(1), 1–22.
Bolton, L. E., & Alba, J. W. (2006). Price fairness: Good and service differences and the
role of vendor costs. Journal of Consumer Research, 33, 258–265.
Bolton, R. N., & Myers, M. B. (2003). Price-based global market segmentation for ser-
vices. Journal of Marketing, 67,108–128.
Caro, L. M., & Garcia, J. A. M. (2007). Consumer satisfaction with a periodic reoccurring
sport event and the moderating effect of motivations. Sport Marketing Quarterly, 16,
70–81.
Dean, A., Morgan, D., & Tan, T. E. (2002). Service quality and customer willingness to
pay more for travel services. Journal of Travel & Tourism Marketing, 12(2), 95–110.
Fischer, B. (2015). Clearing the next obstacle. Street & Smith’s Sports Business Journal,
October 26–November 1, 17.
Greenwell, T. C. (2007). Expectations, industry standards, and customer satisfaction in
the student ticketing process. Sport Marketing Quarterly, 16, 7–14.
Greenwell, T. C., Brownlee, E., Jordan, J. S., & Popp, N. (2008). Service fairness in
spectator sport: The importance of voice and choice on customer satisfaction. Sport
Marketing Quarterly, 17, 71–78.
Han, J., & Kwon, H. H. (2009). The mediating effect of perceived quality between extrin-
sic cues and perceived value in ski products. International Journal of Sports Marketing &
Sponsorship, 10(4), 287–301.
Haws, K. L., & Bearden, W. O. (2006). Dynamic pricing and consumer fairness percep-
tions. Journal of Consumer Research, 33, 304–311.
Helkkula, A., & Kelleher, C. (2010). Circularity of customer service experience and cus-
tomer perceived value. Journal of Customer Behavior, 9(1), 37–53.
Herrmann, A., Xia, L., Monroe, K. B., & Huber, F. (2007). The influence of price fairness
on customer satisfaction: An empirical test in the context of automobile purchases.
Journal of Product and Brand Management, 16(1), 49–58.
Homburg, C., Koshate, N., & Hoyer, W. D. (2005). Do satisfied customers really pay more?
A study of the relationship between customer satisfaction and willingness to pay. Journal
of Marketing, 69, 84–96.
Itani, O.S., Kassar, A.N., & Loureiro, S.M.C. (2019). Value get, value give: The rela-
tionship among perceived value, relationship quality, customer engagement, and value
consciousness. International Journal of Hospitality Management, 80, 78–90.
 Marketing Obstacles Facing Obstacle Racing 197

Kim, J., Natter, M., & Spann, M. (2009). Pay what you want: A new participative pricing
mechanism. Journal of Marketing, 73, 44–58.
Kwak, D. H., McDaniel, S., & Kim, K. T. (2011). Revisiting the satisfaction-loyalty rela-
tionship in the sport video gaming context: The mediating role of consumer expertise.
Journal of Sport Management, 1–43.
Kwon, H. H., Trail, G., & 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.
Malc, D., Mumel, D., & Pisnik, A. (2016). Exploring price fairness perceptions and their
influence on consumer behavior. Journal of Business Research, 69(9), 3693–3697.
McKay, B., & McKay, K. (2020). The history of obstacle courses for military fitness,
sport, and all around toughness. Retrieved on December 11, 2020 from https://www.art
ofmanliness.com/articles/the-history-of-obstacle-courses/#:~:text=But%20obstacle%20
racing%20in%20terms,UK%20called%20the%20Tough%20Guy.&text=Two%20
weeks%20after%20the%20first,first%20Spartan%20Race%20took%20place
Moe, W., & Fader, P. S. (2009). The role of price tiers in advance purchasing of event
tickets. Journal of Service Research, 12(August), 73–86.
Mullin, N. (2012). Obstacle course challenges: History, popularity, performance demands,
effective training, and course design. Journal of Exercise Physiology, 15(2), 100–128.
Murray, D., & Howat, G. (2002). The relationship among service quality, value, satisfac-
tion, and future intentions of customers at an Australian sports and leisure center. Sport
Management Review, 5, 25–43.
Oh, H. (2003). Price fairness and its asymmetric effects on overall price, quality, and value
judgments: The case of an upscale hotel. Tourism Management, 24, 387–399.
Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfac-
tion decisions. Journal of Marketing Research. 17, 460–69.
Oliver, R. L. (1999). Whence customer loyalty. Journal of Marketing, 63, 33–44.
Olsen, L. L., & Johnson, M. D. (2003). Service equity, satisfaction, and loyalty: From
transaction-specific to cumulative evaluations. Journal of Services Research, 5(3), 184–195.
Parasuraman, A. (1997). Reflections on gaining competitive advantage through customer
value. Journal of Academy of Marketing Science, 25(2), 154–161.
Parasuraman, A., & Grewal, D. (2000). The impact of technology on the quality-value-
loyalty chain: A research agenda. Journal of Academy of Marketing Science, 28(1), 168–174.
Park, J., Ellis, G. D., Kim, S. S., & Prideaux, B. (2010). An investigation of perceptions of
social equity and price acceptability judgments for campers in the U.S. national forest.
Tourism Management, 31, 202–212.
Petrick, J. F. (2002). Development of a multi-dimensional scale for measuring the per-
ceived value of a service. Journal of Leisure Research, 34(2), 119–134.
Petrick, J. F. (2004). First timers’ and repeaters’ perceived value. Journal of Travel Research,
43, 29–38.
Petrick, J. F. (2005). Segmenting cruise passengers with price sensitivity. Tourism
Management, 26, 753–762.
Petrick, J. F., Backman, S. J., & Bixler, R. D. (1999). An investigation of selected fac-
tors’ impact on golfer satisfaction and perceived value. Journal of Park and Recreation
Administration, 17(1), 40–59.
Radnedge, C. (2019). Obstacle course racing scaling its way to Olympic inclusion. Reuters
Sports News. Retrieved on December 11, 2020 from https://www.reuters.com/article/us-
sport-ocr/obstacle-course-racing-scaling-its-way-to-olympic-inclusion-idUSKCN1SH08F
198 J. Reese

Rao, A. R. (2005). The quality of price as a quality cue. Journal of Marketing Research, 42,
401–405.
Rascher, D. A., McEvoy, C. D., Nagel, M. S., & Brown, M. T. (2007). Variable ticket
pricing in Major League Baseball. Journal of Sport Management, 21, 407–437.
Rishe, P. J., & Mondello, M. (2003). Ticket price determination in the National Football
Leagues: A quantitative approach. Sport Marketing Quarterly, 12(2), 72–79.
Sivakumar, K. (2000). Price-tier competition: An integrative review. Journal of Product
and Brand Management, 9, 276–290.
Suh, J. C., & Yi, Y. (2006). When brand attitudes affect the customer satisfaction-loyalty
relation: The moderating role of product involvement. Journal of Consumer Psychology,
16(2), 145–155.
Trail, G. T., Anderson, D. F., & Fink, J.S. (2005). Consumer satisfaction and identity the-
ory: A model of sport spectator conative loyalty. Sport Marketing Quarterly, 14, 98–111.
Tsuji, Y., Bennett, G., & Zhang, J. J. (2007). Consumer satisfaction with an action sports
event. Sport Marketing Quarterly, 16, 199–208.
Weedon, G. (2015). Camaraderie reincorporated: Tough Mudder and the extended distri-
bution of the social. Journal of Sport and Social Issues, 39(6), 431–454.
Wilkinson, J. (2009). Customer base management for dynamic markets. International
Journal of Market Research, 51(4), 547–550.
Xia, L., Monroe, K. B., & Cox, J. L. (2004). The price is unfair! A conceptual framework
of price fairness perceptions. Journal of Marketing, 68, 1–15.
Xu, Y., Goedegebuure, R., & van der Heijden, B. (2006). Customer perception, customer
satisfaction, and customer loyalty within Chinese securities business: Towards a medi-
ation model for predicting customer behavior. Journal of Relationship Marketing, 5(4),
79–104.
Zeithaml, V. A. (1988). Consumer perceptions of price, quality and value: A means-end
model and synthesis of evidence. Journal of Marketing, 52, 2–22.
Chapter 11

Congruity Perceptions
in Endurance Event
Participant Consumers
Jason A. Rice, Marion E. Hambrick,
and T. Christopher Greenwell

Introduction
Endurance events in the United States, such as road running marathons/half-
marathons and obstacle course events, play a vital role in the $1.4 billion running
industry (Bershidsky, 2019). Running clubs and specialty stores across the United
States recruit new customers and build rapport through community training
programs designed for a number of these endurance events. Performance brands
(e.g., nutrition, shoes, and apparel) also utilize endurance events as marketing
communication channels. As a result of this far-reaching influence, changes in
endurance event participation trends signal and impact greater industry trends.
The United States endurance event industry has witnessed multiple booms and
contractions over the past quarter-century (Miller, 2017). The most recent growth
period, fueled by the introduction of non-traditional events (e.g., mud runs, color
runs, and obstacle events), has experienced a cooling effect. After record, high
participation numbers in 2013, Running USA (2016) reported contraction in the
market, with all events types experiencing declines over the next few years. Even
in more recent years, participation numbers have not returned to their previous
highs (Running USA, 2018).
While it is unclear if experienced event participants are leaving the market or
the pipeline of new participants has diminished, capturing a better understand-
ing of these consumers is warranted. One way to understand these changes is to
evaluate how individuals perceive these events. Past consumer behavior research
has suggested evaluating choices based on perceptions of their utilitarian brand
features and symbolic meaning (Kressmann et al., 2006). “The match between
a consumer’s ideal expectations of utilitarian brand features and their percep-
tions of how the product is perceived along the same features” has been identified
as functional congruity (Kressmann et al., 2006, p. 955). In the case of endur-
ance event participation, utilitarian features may include providing a challeng-
ing experience, the ability to obtain certain physical health outcomes, or social
opportunities (Masters et al., 1993, 2003; Rice, 2015; Summers et al., 1983).
Symbolic meaning, on the other hand, is interpreted as the perceived fit
between a person’s self-concept with the brand (the event) or brand users (event

DOI: 10.4324/9781003302476-11
200 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

participants), hereby referred to as self-congruity (Aguirre-Rodriguez et al., 2012).

When the brand or brand users fit the consumer’s self-concept, the consumer is
more likely to interact with it. An example of poor fit has been demonstrated
by the African-American runner experience. This group is an under-represented
population in events such as marathons. Research has identified how African-
Americans may view long-distance running as something white, not black, peo-
ple do (Rice et al., 2018), resulting in low levels of self-congruency and impacting
involvement. These brand perceptions may also differ by levels of involvement.
According to the psychological continuum model (PCM), consumers progress
through a series of identifiable levels, from awareness, where the activity has yet
to be undertaken, through allegiance (Funk & James, 2001; Funk et al., 2011;
Iwasaki & Havitz, 1998). Such a progressive model is important for sport market-
ers as it presents an opportunity to segment consumers by involvement profiles.
To date, investigations into the participant sport consumer involvement process
have remained mostly conceptual, with limited research exploring influential fac-
tors facilitating and developing movement into and along the process.
Previous sport participant studies identified certain motivations for those par-
ticipating in endurance events (Masters et al., 1993; Ogles & Masters, 2003; Rice,
2015; Summers et al., 1983) but only categorized their sample as runners, failing
to consider the range of consumer involvement. Most noticeably, the event non-
participant is omitted from their evaluation. Others have explored the relation-
ship between components influencing the sport involvement process (Beaton et al.,
2011) and the effects of including additional components (Ridinger et al., 2012),
but again omit event non-participants and do not evaluate the influence of con-
ceptualized moderators. Building upon the efforts of past research, while under-
standing the value of potential event participants, the purpose of the current
study is to investigate the presence and influence of consumer congruity on their
sport involvement process.
With a declining consumer base, the endurance event market is challenged
to attract new participants to the sport, while also remaining relevant to cur-
rent participants occupying various levels of involvement. Therefore, this study
combines sport activity involvement theory with consumer congruity research
to better understand why consumers engage in and persist in endurance event
participation. Further, this study utilizes a measure of congruity to evaluate dif-
ferent levels of consumers’ involvement. By identifying where endurance events
are either congruent or incongruent with consumers’ expectations, marketers can
better frame events and brand communications for specific consumer groups.

Review of Literature
As United States endurance sport participation numbers have witnessed a down-
ward trend, the industry must establish itself as a relevant option. Research
has found event consumers/participants are influenced by a variety of func-
tional attributes. Individuals expect endurance events to provide a sense of fun
 Congruity Perceptions 201

or enjoyment, an opportunity to complete or obtain a sense of achievement, a

social atmosphere, and assistance with improving health (Masters et al., 1993;
McDonald et al., 2002; Ogles & Masters, 2000; Rice, 2015; Ridinger et al., 2012;
Summers et al., 1982, 1983). When consumers perceive endurance events as deliv-
ering these functional attributes, they will increase their involvement in them
(Kressmann et al., 2006). While functional attribute evaluation is considered to
influence a consumer’s involvement with a brand, research has also suggested
more symbolic measures, here called measures of self-congruity, have as much,
if not greater, influence on the consumer’s involvement process (Kressmann
et al., 2006). Still, limited research on self-congruity’s role in consumer decisions
has been conducted in the sport context (Horton & Mack, 2000; Lough et al.,
2014). This lack of research is surprising as many researchers have conceptual-
ized the existence of self-congruency in the sport involvement process (Funk &
James, 2001; Funk et al., 2011; Havitz & Dimanche, 1997). The current research
addresses this void and follows the suggestion of Beaton et al. (2011), who recom-
mended efforts exploring factors that facilitate the development and progression
of involvement continue.

The Psychological Continuum Model (PCM)

Runners, like other sport consumers, exhibit a variety of characteristics, needs,
motivations, and goals. As such, it would be unwise to treat all runners as the
same, as participants in this activity tend to exhibit a broad range of levels of
involvement. The PCM provides a basis for examining runners at different levels.
The PCM was designed to evaluate an individual’s level of involvement with a
sport object, such as level of sport fandom, and later has been applied to sport and
leisure activity participation (Funk & James, 2001). A key attribute of the PCM is
its focus on the social-psychological connection an individual makes with a sport
object, participation in an endurance event in the case of this study. The PCM
identified that high levels of sport involvement exist when individuals evaluate
their participation in a sport activity as a central component of their life, provid-
ing both hedonic and symbolic value, measured as centrality, pleasure, and sign
(Beaton et al., 2011). As a level-based framework, the PCM examines the socio-
logical and psychological processes associated with attitude formation along four
vertical levels: (a) awareness, (b) attraction, (c) attachment, and (d) allegiance
(Funk & James, 2001; Funk et al., 2011).
An initial evaluation of potential participation is triggered from external influ-
ences interacting with and shaping perceived and desired identities. This eval-
uation may come in the form of matching activities with perceived or desired
identities held by the individual or other functional benefits (Beaton et al., 2011).
The strength of a consumer’s commitment is determined by an assignment of
emotional, functional, and symbolic meaning to the activity (Funk & James,
2006) and by the extent to which the consumer identifies with important val-
ues and self-images associated with participation (McDonald et al., 2002). Once
202 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

participation in the event is perceived as properly aligned with self-concept, a

level of commitment and attitudinal loyalty is formed, leading to repeat behavior
(Meyer et al., 2004). While it is understood that individuals with no previous
endurance event participation may occupy the awareness level of this model, the
classifying characteristics sport marketers can use to effectively identify this seg-
ment beyond participation has been missing from sport participation research.
The current research takes the call for more evaluation exploring factors facili-
tating the development and progression of involvement (Beaton et al., 2011) by
providing an analysis of influences and facilitators conceptualized within the
involvement process, here identified as congruity.

Consumer Congruency
Congruency is important in understanding consumer behavior because of the
associations consumers hold about brands and the role these meanings take in
the decision process. Various types of congruency have been acknowledged in
past consumer research with two main categories, functional congruity and self-
congruity. Functional congruity is “the match between a consumer’s ideal expec-
tations of utilitarian brand features and their perceptions of how the product is
perceived along the same features” (Kressmann et al., 2006, p. 955). In a sport par-
ticipation context, the influence of functional attributes on the involvement pro-
cess has been identified in several studies (Funk et al., 2011; Masters et al., 1993;
McDonald et al., 2002; Ogles & Masters, 2000; Rice, 2015; Ridinger et al., 2012;
Scanlan et al., 2003; Summers et al., 1982; 1983). Functional expectations from
event participation have included feeling a sense of fun or enjoyment, having
the ability to compete, receiving socialization opportunities, improving health,
and experiencing a sense of achievement. Functional congruity is most impactful
when a participant desires the opportunities a brand category is expected to offer,
and an event is perceived as able to deliver them.
The more symbolic form of congruity, self-congruity, is an internal process
driven by individuals matching their self-concept with a target. Perceptions of
self include both actual self-concept and ideal self-concept. The actual self
is how consumers truly see themselves (Sirgy & Su, 2000), while the ideal self is
how consumers would like to be (Hung & Petrick, 2011). In the case of endurance
event participants, self-concept can range from self-identified non-runners to very
serious runners. For example, someone driving a vehicle with multiple Ironman
event or 26.2 stickers is demonstrating themselves as an endurance event athlete.
Conversely, someone identified as a non-athlete would not demonstrate this and
respond negatively when asked if they are a runner. With ideal self, someone not
currently self-identified as a runner could harbor desires to be a runner. And so,
actual and ideal self-concepts can differ.
Once self-concept is formed, they are evaluated against a target. Previous con-
gruency research has identified that the target evaluated may be interpreted as
a brand (Aguirre-Rodriguez et al., 2012) or a “stereotypical image of a person
 Congruity Perceptions 203

believed to be found in a certain situation” (Kang, 2002, p. 174). Sport consumer

research has identified the product as a sport category, such as basketball, and the
brand as the team (Pritchard and Funk, 2010). The current research follows suit by
identifying the product as running and the brand as the running event of interest,
and so, the evaluation of congruity occurs between self-image and the brand.
This is possible because consumers hold symbolic meanings and associations of
brands that are evaluated during the purchase decision process (Gladden et al.,
1998; Ross, 2006, 2008). In the case of endurance events, research has shown that
symbolic meanings held for endurance events may include demanding, physical,
committed, brave, insane, fun, social, and rewarding. Past research has suggested
evaluating both functional and symbolic self-congruity for a complete measure
of a consumer’s decision-making process (Hung & Petrick, 2011). For this reason,
the current study identified and measured one evaluation of functional congruity
and three forms of self-congruity; brand association congruity, a stereotypical user
with ideal self-image congruity, and a stereotypical user with actual self-image
congruity. These four forms of congruity are defined within the context of this
study in Table 11.1.
Congruency theory is a natural fit with popular sport involvement models and
should assist in producing beneficial practical outcomes. As a level-based con-
struct, the PCM relies on transitions between four different levels of involvement.
As previously discussed, it has been hypothesized that each level has its own
set of inputs/antecedents and outcomes/characteristics which initiate process-
ing between the levels (Beaton et al., 2011). Conceptualizations behind congru-
ency theory are represented in processes identified within all four of these levels.
During the initial evaluation at the awareness level, a non-runner’s attitudes are
shaped by external influences such as social and cultural beliefs, media, family,
and peers (Beaton et al., 2009), all agents that influence the creation of self-image

Table 11.1 Congruity Definitions

Congruity Type Definition

Functional congruity An evaluation of ideal utilitarian expectations of endurance

event participation and an individual’s perception of how
long-distance road running participation is perceived
along with the same attributes.
Brand association congruity The match between an individual’s actual self-image with
perceived associations of long-distance road running
events.
Stereotypical user with ideal The match between how an individual would like to be
self-image congruity (ideal self-image) with the image of a person believed to
be a typical long-distance road running event participant.
Stereotypical user with actual The match between who an individual really thinks they
self-image congruity are (actual self-image) with the image of a person
believed to be a typical long-distance road running event
participant.
204 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

used in congruency theory (Tajfel & Turner, 2010). These event non-participants
are likely to shape their actual and ideal self-image in response to the target. As
individuals move from awareness to actual participation in the activity, identified
as the next stage, attraction, and on to further levels, they continue to process
information about the brand by evaluating its ability to satisfy functional and
symbolic needs (Beaton et al., 2009; Funk & James, 2006).
Examination of the PCM confirms the theoretical placement of congruency
theory within an evaluation of sport involvement. In response to participation
trends within the United States endurance event market, the current research
evaluated a range of consumers, from non-endurance event participants to serious
runners, segmenting them based on involvement level. Characteristics of level
occupants, based on the theoretical underpinning of sport involvement research,
demonstrate the presence of consumer congruity within sport involvement the-
ory. In other words, runners with higher levels of involvement should perceive the
activity to be more congruous with their view and place in the world. As a result,
the following hypotheses were generated.

H1: Functional congruity will increase as the level of involvement increases.

H2: Brand association congruity will increase as the level of involvement
increases.
H3: Stereotypical user with ideal self-image congruity will increase as the
level of involvement increases.
H4: Stereotypical user with actual self-image congruity will increase as the
level of involvement increases.

Method

Participants
Understanding the need to maintain current event participants while attract-
ing new participants, this study investigated a range of involvement levels, from
non-endurance event participants to highly involved endurance event partici-
pants. The sample was recruited from 11 endurance training groups located
throughout the United States. Training groups were chosen because many of
these programs are designed to prepare individuals for endurance event participa-
tion. Often, the culminating activity for a group is a specific event or a collection
of endurance events of different distances. This sample was considered appro-
priate due to the insight they would have toward endurance events and event
participation through a range of involvement levels.
The sample also included students from two undergraduate and one graduate
sport management courses at a public university in the United States Midwest.
A sample of college students was determined adequate for several reasons. First,
college students are endurance event participants. It is also a time when individ-
uals are exposed to endurance event participation (Meschke, 2018; Ruiz, 2019).
 Congruity Perceptions 205

Next, previous brand association research has utilized this type of sample (Aaker,
1997; Braunstein & Ross, 2010; Lee & Cho, 2009; Ross, 2008). Finally, student
samples are often used in sport consumer researcher because they are significant
consumers of sport and reflect a valued demographic for marketers (Braunstein &
Ross, 2010; Ross, 2008).
In total, 650 questionnaires were collected, with 42 containing missing
responses to key demographic or variable items. These were removed from the
analysis. Of those remaining, 541 indicated respondents had previously partici-
pated in an endurance event (marathon or half marathon), and 67 indicated no
previous endurance event participation experience but were familiar with endur-
ance events. Following Funk and James (2006) and Beaton et al. (2009; 2011),
these 67 respondents were automatically placed in the PCM’s awareness level. Those
with previous endurance event participation were placed into involvement levels
based on their responses. The current study utilized a staging mechanism devel-
oped by Beaton et al. (2009), creating ranked involvement profiles from scores
of three facets of involvement; pleasure, sign, and centrality. The collection
of possible involvement profiles was then translated into a stage algorithm with
endurance event participant respondents placed into one of the remaining three
involvement levels. After this staging, 326 individuals occupied the highest level of
allegiance, followed by 174 respondents in attachment, 42 in attraction, and 63
in the lowest level of awareness. The demographic profile of the endurance event
sample was considered representative based on its similarity to that of endurance
event participants in the United States. Running USA (2017) estimates the United
States has more female runners than male, with 63% female runners, and the largest
category of runners includes 35- to 44-year-olds (Running USA, 2017). A full report
of demographic characteristics for the sample can be found in Table 11.2.

Measures
Each respondent received the same questionnaire consisting of three sections.
The first section was comprised of qualifying items to assist with the remainder
of the survey. This included the history of event participation. If respondents
had not previously participated in an event, they were asked if they knew what
an endurance event was. Those with no knowledge of endurance events were
finished with the survey and excluded from the results. Those with no participa-
tion history but with knowledge of endurance events were retained and received
questions measuring all constructs represented in the second section of the ques-
tionnaire except involvement’s pleasure, sign, and centrality. Those indicating
previous event participation received all items in the second section.
Within the second section of the questionnaire, the following constructs were
measured: (a) involvement, (b) functional congruity, (c) stereotypical user con-
gruity, and (d) brand association congruity with traditional endurance events
(long-distance running events such as half-marathons and marathons). Past
research has indicated non-traditional and traditional endurance events have
206 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

Table 11.2 Sample Characteristics

Variable % Total

Sex
Male 25% 154
Female 74% 447
No Response <1% 7
Age
18–24 11% 63
25–34 27% 165
35–44 30% 184
45–54 23% 137
55–64 7% 45
65+ 1% 6
No response 1% 8
Ethnicitya
White (non-Hispanic) 82% 524
Black (non-Hispanic) 6% 37
Hispanic or Latino 5% 29
Asian or Pacific Islander 3% 16
American Indian or Alaskan Native 1% 5
Other 1% 6
Would rather not answer 1% 7
No response 1% 6
Past Event Participationa
Marathon 49% 315
Half Marathon 83% 528
Obstacle Course Event 32% 203
No Previous Endurance Event Participation 10% 64
PCM Level for Traditional Endurance Events
Allegiance 54% 326
Attachment 29% 174
Attraction 7% 41
Awareness 11% 67
a Respondents could choose all that apply. Percentage total >100%.

distinctive brand characteristics (Rice, 2015), and the current analysis would ben-
efit from a clear evaluation of one category. While non-traditional events have an
impact, as witnessed by their influence on recent trends, traditional endurance
events remain the pillar of the industry, and their long history is suggested to
produce a more stable influence. For these reasons, an evaluation of traditional
endurance events was considered.
As suggested by Beaton et al. (2011), involvement was measured with nine
items (three for pleasure, sign, and centrality) on a 7-point Likert-type scale (1 =
strongly disagree, 7 = strongly agree). Each of the scale scores was used to calcu-
late the occupancy of an involvement previously described (Beaton et al., 2009).
Functional congruity, the ability to fulfill the perceived utilitarian attributes gen-
erally expected with endurance event activity participation, was measured using
 Congruity Perceptions 207

an ideal-point model based on the research of Kressman et al. (2006). An exhaus-

tive literature review found specific functional attributes related to endurance
sport participation include obtaining a sense of achievement, having a fun expe-
rience, developing camaraderie, receiving an opportunity to be healthy, and sat-
isfying a need to compete (Funk et al., 2011; Masters et al., 1993; McDonald et al.,
2002; Ogles & Masters, 2000, 2003; Rice, 2015; Ridinger et al., 2012; Scanlan
et al., 2003, 2013; Summers et al., 1982, 1983). The functional congruity score
was determined by calculating the direct score of the five functional associations
of endurance event participation by asking if “Long-distance road running event
training and participation [provide the functional attribute]”. Scores were calcu-
lated on a 7-point Likert-type scale (1 = strongly disagree, 7 = strongly agree). To
produce a total functional congruity score, a mean score from these five different
scores was calculated.
The congruency with the stereotypical user was measured on a 5-point Likert-
type scale (1 = not at all overlapped, 5 = nearly total overlap) and (1 = not at all sim-
ilar, 5 = nearly total similarity) based on Kang (2002) and Kwak and Kang (2009).
Respondents were asked to “Please take a minute to think about the overall image
of a typical long-distance road running event participant and respond to the fol-
lowing questions”. With these pictures in mind, they were asked to respond to the
items identifying their actual-self congruity and ideal-self congruity with a stereo-
typical user. For example, actual self-image was measured with the following item,
“How much does your own actual self-image (who you think you actually are) and
the perceived overall image of the typical long-distance road running event par-
ticipant overlap?” Past studies demonstrated reliable and valid scores using similar
measures. In Kang (2002), university students were asked to evaluate measures
of self-image congruity for two distinct physical activities. Overall model fit was
acceptable, and factor loadings for both physical activities were high. To produce
a stereotypical user congruity score for both actual and ideal self-image, the mean
from each set of items for their individual measure of the stereotypical user was
calculated.
The researchers were also interested in the effect self-image congruity had on
the perceived associations of a category brand. This is conceptualized as brand
association congruity. To overcome issues identified by Ross (2008) and Caprara,
Barbaranelli, and Guido (2001) with a popular 15-item brand association scale
developed by Aaker (1997), the current measure of brand association congru-
ity utilized a direct measure of congruence, as opposed to the difference score
technique. This measure is supported by the work of Govers and Schoormans
(2005) and Sirgy et al. (1997). The current analysis asked all respondents to
“List the characteristics which first come to mind when thinking about a typical
long-distance road running event”. To receive the direct congruency measure, the
questionnaire then requested respondents to consider the associations they just
provided when answering the following three items on a 7-point Likert-type scale
(1 = strongly disagree, 7 = strongly agree): “A typical long-distance road running
event is like me”, “I identify with my description of a typical long-distance road
208 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

running event”, and “Considering my own personality and comparing it to the

description I just provided for a typical long-distance road running event, I find
they are similar”. A total self-image congruity score was calculated by producing
the mean score from the three items.
The final section asked respondents to record their basic demographic infor-
mation. This included age, sex, marital status, race, number of years participat-
ing in endurance events, and previous race participation behavior over the past
12 months. These data were used for descriptive purposes to determine if the
sample was representative to endurance event participant profiles, as indicated by
Running USA (2017), and to allow for further analysis.

Data Analyses
All analyses in this study were conducted using the Statistical Package for the
Social Sciences (SPSS 16.0). Specifically, descriptive statistics, multivariate anal-
ysis of variance tests (MANOVA), and Cronbach’s alpha coefficients (α) were
used to investigate the role of congruity within the participant sport involvement
process. The treatment of the data was separated into three parts. First, descrip-
tive statistics, including the means and standard deviations of all variables and
involvement PCM level assignments for respondents, were calculated. Second,
the reliability of the involvement scores and congruity scores were assessed by
examining their Cronbach’s alphas, as recommended by Nunnally and Bernstein
(1994). Next, MANOVA was utilized to examine the relationship between par-
ticipant PCM involvement level and the role of congruity. To examine the dif-
ferences in the congruity effect between the four PCM involvement levels, the
MANOVA utilized a grouping variable, PCM level, as the independent variable.
The four types of congruity (functional congruity, stereotypical user congruity
with actual self, stereotypical user congruity with ideal self, and brand association
congruity) served as the dependent variables. These findings could determine
which types of congruity are most associated with each participant’s involvement
level. Because the study was exploratory interested in finding and highlighting
differences, a univariate post hoc analysis was chosen.

Results
Four distinct types of congruity with endurance events were measured in the cur-
rent analysis to evaluate the utlilitarian and symbolic relationships an individual
had with endurance running and events. These results are displayed in Table 11.3.
Confirmatory factor analysis indicated a good fit of the model to the data (CMIN/
DF = 1.604, CFI = .983, GFI = .934, AGFI = .906, RMSEA = .045). As expected,
results indicated the measures of congruity did increase as the level of involve-
ment increases. Results from the MANOVA test revealed significant differences
among the four unique types of congruity scores between the different levels of
endurance event involvement, where Wilks’ Lambda = 0.6281, F(4, 593) = 28.862,
 Congruity Perceptions 209

Table 11.3 Descriptive Statistics for Univariate Analysis

Stereotypical User
Brand Stereotypical User with Actual
Association with Ideal Self- Self-Image
Functional Congruity Congruity Image Congruity Congruity
Involvement
Level Mean SD Mean SD Mean SD Mean SD
Allegiance 6.65 1 0.59 5.74 1,2,3 1.24 3.731,2,3 0.82 3.27 1,2,3 0.86
Attachment 6.441 0.54 5.181,4 1.32 3.331,4 0.91 2.851,4 0.79
Attraction 6.341 0.56 4.611,4 1.36 3.114 0.81 2.594 0.79
Awareness 5.082,3,4 1.44 3.402,3,4 1.94 2.733,4 1.11 2.353,4 0.99
1 p < 0.01 with awareness.
2 p < 0.01 with attraction.
3 p < 0.01 with attachment.
4 p < 0.01 with allegiance.

p < 0.001. The follow-up between-groups univariate tests revealed multiple signif-
icant differences between the PCM levels and measures of congruity.
To address hypothesis 1, an analysis of functional congruity, respondents in the
allegiance (m = 6.65), attachment (m = 6.44), and attraction (m = 6.34) groups
reported significantly higher functional congruity than those in the awareness
group (m = 5.08). For hypothesis 2, which addressed brand association congruity,
respondents in the allegiance group reported significantly higher brand association
congruity (m = 5.74) than respondents in the other three groups. Respondents in
the attachment (m = 5.18) and attraction (m = 4.61) groups were similar but signif-
icantly higher than respondents in the awareness group (m = 3.40). Hypothesis 3,
concerned with a stereotypical user with ideal self-image congruity, respondents
in the allegiance group reported significantly higher stereotypical user with ideal
self-image congruity (m = 3.73) than the other three groups. Respondents in the
attachment group (m = 3.33) were significantly higher than those in the aware-
ness group (m = 2.73). And finally, in evaluating for hypothesis 4, respondents in
the allegiance group reported significantly higher stereotypical users with ideal
self-image congruity (m = 3.27) than the other three groups. Respondents in the
attachment group (m = 2.85) were significantly higher than those in the aware-
ness group (m = 2.25).
The greatest differences in congruity measures between the involvement levels
were recorded between the awareness and allegiances levels. For functional con-
gruity, the scores for awareness level occupants were significantly different from
all other involvement levels. This was an effect of the large difference between
awareness (m = 5.08) and attraction (m = 6.34), with scores of functional congru-
ity increasing at each consecutive level. Interestingly, no significant differences
were shown between the other involvement levels. This isolated difference was
unique to functional congruity as other congruity measures differed between mul-
tiple involvement levels. For example, the allegiance level experienced significant
210 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

differences with all other involvement levels under each symbolic measure of con-
gruity. However, no significant differences existed between attraction and attach-
ment along with any congruity measure.

Discussion
The current research was designed to further investigate involvement within the
endurance event market by introducing measures of congruity to operationalize
influencers and moderators conceptualized within a popular sport involvement
model. The research was also a response to recent endurance market trends of
stagnation following a rapid growth period. Results revealed differences in con-
gruity measures between some involvement levels, providing valuable insight into
unique consumer segments for sport marketers and the sport involvement process
but also challenged some conceptualizations of the PCM.

Endurance Event Involvement and Congruity

Initial results describe differences in the four measures of congruity (functional,
brand association, stereotypical user with ideal self-image, and stereotypical
user with actual self-image) throughout the involvement levels. Further anal-
ysis revealed that all congruity scores increased in coordination with involve-
ment level, and noteworthy differences existed between many but not all levels.
Specifically, those within the highest involvement level, allegiance, displayed sig-
nificant differences between all other levels along with the three symbolic con-
gruity scores but not with functional congruity. The lowest level of involvement,
awareness, also produced multiple significantly different results, while no signifi-
cant differences existed between those occupying the middle involvement levels,
attraction, and attachment within each congruity measure. As conceptualized,
differences exist between the two main categories of congruity. Current results
reveal functional and symbolic congruity are interpreted differently within the
endurance event market. Even within symbolic measures, brand association and
the two stereotypical user congruity measures are also distinct. These results are
in line with past research (Parker, 2009) comparing different congruity measures.
Functional congruity scores were the highest among all congruity measures.
Non-endurance event participants occupy the awareness level, yet believe endur-
ance events deliver desirable outcomes yet choose to not participate, it could
be these outcomes are not considered desirable by these individuals, or in the
least, are not motivated enough to begin participation. Still, the only signifi-
cant increase in functional congruity occurred between this awareness level and
attraction before a relative plateauing of scores through the next two involve-
ment levels. This suggests that even though event participation is expected by
many to deliver certain outcomes, a significant shift occurs as event participa-
tion commences. As a result, this increased emphasis on functional outcomes
could be a trigger between event participation and non-participation but does
 Congruity Perceptions 211

not dramatically increase involvement once participation is initiated. This is sup-

ported in the conceptualization of the PCM, which suggests sport participants
move from valuing utilitarian outcomes to more personal and symbolic measures
(Beaton et al., 2011).
The symbolic measures of congruity deliver important results in how scores
can be distinguished between various involvement levels. While the symbolic
congruity measures did increase along all PCM involvement levels, a categoriza-
tion of three involvement levels are suggested by the results. The awareness and
allegiant levels remain consistent with the PCM, scoring the lowest and highest
measures, but the mid-levels of attachment and attraction recorded insignificant
differences. These results suggest symbolic congruity measures have the ability
to distinguish non-event participants and highly involved individuals from other
levels.
Other unique results include the difference between actual self-image congruity
with a stereotypical user and the higher scoring ideal self-image congruity. This
implies a stereotypical endurance event participant is closer to the questionnaire
respondent’s ideal self than actual self and establishes a difference between the
two measures. A subtle demonstration of symbolic congruity can be witnessed
on marathon weekends and Ironman triathlon finish lines. Anyone visiting New
York City over the first weekend in November will witness marathon finishers
proudly wearing their medals in restaurants, walking around town, or at airport
security stations. At the New York City Marathon, this practice is more tradi-
tional than an effort from the marketing team, but other marathons will ask their
finishers over public address or social media announcements to proudly wear their
medals around town or post photos with their medals on the social medal. The
idea is runners should be proud of their runner identity, and public display of such
will only reinforce this identity.
While the current research was conducted on endurance events such as mar-
athons and half-marathons, the Ironman triathlon series provides another great
example of advancing self-image congruity with one of its signature event features.
At the finish line of every event, organizers announce “YOU are an IRONMAN”
to every finisher. For some repeat finishers, this may be a mere ceremony, but for
others, first-time finishers, in particular, this is a profound moment they have
trained very hard to earn. As they join the Ironman club, it is quite possible they
are also reducing the distance between ideal and actual self.

Theoretical Implications
Theoretically, these results confirm the presence of congruity within a sport
participant’s involvement process (Beaton et al., 2009). However, the unique-
ness between each level is not evident. Major changes were present between the
awareness and attraction phase as individuals moved from non-participation to
participation in the target activity. While functional congruity is rated high for
awareness occupants, a significant change in perception still occurs as they move
212 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

from the level of attraction. Since non-event participants exist in the aware-
ness level, the increase in functional attributes is significant at they experience
event participation. Endurance event marketers should continue to rely on these
event attributes in their marketing communications targeted to those with lower
involvement levels.
While significant increases in functional congruity occurred between these
lowest involvement levels, significant differences were not present between other
involvement levels. These results challenge Beaton et al.’s (2009) role of func-
tional meaning in assisting movement between the attachment level to alle-
giance. Instead, the influence of this moderator is determined to exist between
the awareness and attraction levels. Beaton et al.’s (2009) conceptualization of
self-concept’s role between attraction and attachment is also questioned. Current
study results reveal no significant differences in three symbolic measures of
self-congruity between the attraction and attachment levels, while significant
differences exist between the allegiance level and all other levels. While past
research has conceptualized and measured four unique involvement levels (Filo
et al., 2011; Funk & James, 2001; Funk et al., 2011), the current results had diffi-
culty creating distinguishing profiles of these four levels and suggests more inves-
tigation is warranted.

Practical Implications
Because all measures of congruity increase along with the involvement lev-
els with significant increases from awareness to higher involvement levels, race
organizers are encouraged to target not only current participants but also non-
participants with messages containing characteristics suggested by the current
measure of congruity. The New York City Marathon’s social media accounts provide
a good illustration of the event attempting to promote the marathon to a wide
variety of people. Signaling evidence of symbolic congruity, a scroll through their
Instagram account reveals a mosaic of runners. In the images, event participants
represent numerous countries across the world, with unique ethnic backgrounds,
diverse ages from young to old, an array of physical disabilities and body types,
elite runners to first-timers, and even walkers. The diversity of this imagery is pur-
poseful. The New York City Marathon organizers feel their event is for everyone,
and they want everyone to imagine they have a place at their event. The current
research supports these efforts, as lower involved individuals will transition to
higher levels of congruity in relation to increases in involvement levels.
While conceptualizing a match with typical long-distance road running event
participant may be more natural than doing the same with event brand associ-
ations, event producers would benefit from building distinct characteristics for
their event. These may come in the more utilitarian form of functional con-
gruity or the symbolic brand association congruity. Road running events like
Bay to Breakers in San Francisco is known to have a party atmosphere, while a
flat marathon course like the Chicago Marathon is considered to be fast, giving
 Congruity Perceptions 213

participants a chance to accomplish personal records. While the current research

did not distinguish which event characteristics were favored by the different
involvement levels it did signify distinct changes in functional congruity scores
between the awareness level and others.
Regarding brand association congruity, not only could events advertise them-
selves as fun, like Bay to Breakers, but also highlight other event characteristics
an individual could match with themselves. For example, the Walt Disney World
Marathon field contains a lot of first-time marathoners. A race advertising itself
as first-time friendly with lots of runner support along the course would match
well with first-time participants. However, a trail marathon with limited runner
support (participants either carry nutritional needs with them or recruit a per-
sonal support crew) is a better match for a seasoned runner. A variety of event
types can be successful, as there are a variety of runners to attract. The key is for
the road running event market to produce and advertise events with a variety
of characteristics that appeal to and attract the different involvement levels. As
illustrated in the examples, characteristics found within the different measures of
congruity may be used within marketing communications to activate occupants
of the different involvement levels.

Limitations and Future Research

The current study was designed to investigate the process individuals undergo while
becoming involved in a participant sport activity and to evaluate the presence of
congruity within this process. While results produced insight into this process, mul-
tiple research opportunities are presented as extensions of the current study. These
include further investigation of the involvement process, especially occupants of
the awareness level and the influence of congruity measures on consumers. Further
investigation should focus on the involvement process of sport event participants
over time. While the current study provides additional information about the pro-
cess and the presence of congruity measures, its cross-sectional nature limits the
ability to fully understand triggers for level movement. This analysis would be aided
by longitudinal studies focused on identifying why participants move between lev-
els. Results from these studies may help complete our understanding of the involve-
ment process and produce multiple practical benefits.
Involvement levels should also be evaluated in terms of practical outcomes,
such as purchasing behavior and social influence, and theoretical considerations.
The current study revealed attitudinal and behavioral increases in accordance
with involvement levels. However, future analysis could measure the relationships
between the involvement levels and multiple consumption behaviors to further
describe their differences. Behaviors could include direct event consumption such
as when the individual typically registers for an event (early discounted registra-
tion versus waiting until the last minute) or complimentary product consump-
tion such as purchasing running magazines, books, training programs, apparel,
or equipment. Other helpful descriptions of the involvement levels, such as
214 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

likelihood to recommend the event to a friend or possibility of volunteering at an

event, could assist practitioners and further our understanding of the involvement
process and behaviors of segments built around the involvement levels.
The current research found potential issues with the prescribed staging of
involvement levels. Individuals occupying the lowest involvement levels, aware-
ness, and attraction seem to be arbitrarily placed in these levels because they have
or have not participated in the activity. Within the current study, some occu-
pants of the attraction level would have fallen into the awareness level because
they scored low in the three facets of involvement: pleasure, sign, and central-
ity. However, because they had previously participated in endurance events,
they automatically fell into the attraction level based on the staging mechanism
presented by Beaton et al. (2009). Similarly, limited psychological changes, here
conceptualized as congruity, between mid-range involvement levels were pres-
ent. Further investigation into these involvement levels is needed to appropri-
ately assign and describe level occupants. A limitation of the current study is the
sample of college students. Because the psychological development of younger
consumers varies from their older counterpart, the study results representing the
students may be generalized with caution.

References
Aaker, J. L. (1997), “Dimensions of brand personality”, Journal of Marketing Research,
Vol. 34 No.3, pp. 347–356.
Aguirre-Rodriguez, A., Bosnjak, M., & Sirgy, M. J. (2012), “Moderators of the self-congruity
effect on consumer decision-making: A meta-analysis”, Journal of Business Research,
Vol. 65 No. 8, pp. 1179–1188.
Beaton, A., Funk, D., & Alexandris, K. (2009), “Operationalizing a theory of participa-
tion in physically active leisure”, Journal of Leisure Research Vol. 41 No.2, pp. 177–203.
Beaton, A., Funk, D. C., Ridinger, L., & Jordan, J. (2011), “Sport involvement: A con-
ceptual and empirical analysis”, Sport Management Review, Vol. 14 No. 2, pp. 126–140.
Bershidsky, L. (2019), “Why is running on the decline?” Bloomberg, available at: https://
www.bloomberg.com/opinion/articles/2019-06-15/fewer-and-slower-the-decline-of-
running (accessed July 18, 2019).
Braunstein, J. R., & Ross, S. D. (2010), “Brand personality in sport: Dimension analysis
and general scale development”, Sport Marketing Quarterly, Vol. 19 No. 1, pp. 8–16.
Caprara, G., Barbaranelli, C., Guido, G. (2001), “Brand personality: How to make the
metaphor fit?” Journal of Economic Psychology, Vol. 22 No. 3, pp. 377–395.
Filo, K., Funk, D., & O’Brien, D. (2011), “Examining motivation for charity sport event
participation: A comparison of recreation-based and charity-based motives”, Journal of
Leisure Research, Vol. 43 No. 4, pp. 491–518.
Funk, D., Beaton, A., & Pritchard, M. (2011), “The stage-based development of physi-
cally active leisure: A recreational golf context”, Journal of Leisure Research, Vol. 43
No. 2, pp. 268–289.
Funk, D., & James, J. (2001), “The Psychological Continuum Model: A conceptual
framework for understanding an individual’s psychological connection to sport”, Sport
Management Review, Vol. 4 No.2, pp. 119–150.
 Congruity Perceptions 215

Funk, D., & James, J. (2006), “Consumer loyalty: The meaning of attachment in the
development of sport team allegiance”, Journal of Sport Management, Vol. 20 No. 2,
pp. 189–217.
Gladden, J. M., Milne, G. R., & Sutton, W. A. (1998), “A conceptual framework for
assessing brand equity in Division I college athletics”, Journal of Sport Management,
Vol 12, pp. 1–19.
Govers, P. C., & Schoormans, J. P. (2005), “Product personality and its influence on con-
sumer preference”, The Journal of Consumer, Vol. 22 No. 4, pp. 189–197
Havitz, M. E., & Dimanche, F. (1997), “Leisure involvement revisited: Conceptual
conundrums and measurement advances”, Journal of Leisure Research, Vol. 29 No. 3,
pp. 245–278.
Horton, R., & Mack, D. (2000), “Athletic identity in marathon runners; Functional focus
or dysfunctional commitment?”, Journal of Sport Behavior, Vol 23, pp. 101–119.
Hung, K., & Petrick, J. F. (2011), “The role of self and functional congruity in cruising
intentions”, Journal of Travel Research, Vol. 50 No. 1, pp. 100–112
Iwasaki, Y., & Havitz, M. E. (1998), “A path analytic model of the relationships between
involvement, psychological commitment, and loyalty”, Journal of Leisure Behavior, Vol. 30
No. 2, pp. 256–280.
Iwasaki, Y., & Havitz, M. E. (2004), “Examining relationships between leisure involve-
ment, psychological commitment and loyalty to a recreation agency”, Journal of Leisure
Research, Vol. 36 No. 1, pp. 45–72.
Kang, J. (2002), “A structural model of image-based and utilitarian decision-making
processes for participant sport consumption”, Journal of Sport Management, Vol. 16,
pp. 173–189.
Kressmann, F., Sirgy, M. J., Herrmann, A., Huber, F., Huber, S., & Lee, D. (2006), “Direct
and indirect effects of self-image congruence on brand loyalty”, Journal of Business
Research, Vol. 59 No. 9, pp. 955–964.
Kwak, D. H., & Kang, J. (2009), “Symbolic purchase in sport: the roles of self-image con-
gruence and perceived quality”, Management Decision, Vol. 47 No. 1, pp.85–99.
Lee, H. S., & Cho, C. H. (2009), “The matching effect of brand and sporting event person-
ality: Sponsorship implications”, Journal of Sport Management, Vol. 23 No. 1, pp. 41–64.
Lough, N. L., Pharr, J. R., and Owen, J. O. (2014), “Runner identify and sponsorship:
Evaluating the Rock ‘n’ Roll Marathon”, Sport Marketing Quarterly, Vol. 23, pp. 198–211
Masters, K. S., Ogles, B. M., & Jolton, J. A. (1993), “The development of an instrument
to measure motivation for marathon running: The Motivations of Marathoners Scales
(MOMS)”, Research Quarterly for Exercise and Sport, Vol. 64 No. 2, pp. 134–143.
McDonald, M. A., Milne, G. R., & Hong, J. (2002), “Motivational factors for evaluating
sport spectator and participant markets”, Sport Marketing Quarterly, Vol. 11 No. 2,
pp. 100–113.
Meschke, J. (2018), “This college makes every student run a half marathon”, Runners
World, available at https://www.runnersworld.com/runners-stories/a25323961/highlands-
college-all-students-run-half-marathon/, (accessed July 18, 2019).
Meyer, J. P., Becker, T. E., & Vandenberghe, C. (2004), “Employee commitment and moti-
vation: A conceptual analysis and integrative model”, Journal of Applied Psychology,
Vol. 89 No. 6, pp. 991–1007.
Miller, J. A. (2017, November 5), “The running bubble has popped”, New York Times,
available at https://www.nytimes.com/2017/11/05/sports/ny-marathon-running.html,
(accessed July 18, 2019).
216 J. A. Rice, M. E. Hambrick, and T. C. Greenwell

Mullin, B. J., Hardy, S., & Sutton, W. A. (2014), Sport marketing, Human Kinetics,
Champaign, IL.
Nunnally, J. C., & Bernstein, I. H. (1994), Psychometric theory (3rd ed.), McGraw-Hill,
New York, NY.
Ogles, B. M., & Masters, K. S. (2000), “Older vs. younger adult male marathon runners:
Participative motives and training habits”, Journal of Sport Behavior, Vol. 23 No. 2,
pp. 130–143.
Ogles, B. M., & Masters, K. S. (2003), “A typology of marathon runners based on cluster
analysis of motivations.” Journal of Sport Behavior, Vol. 26 No. 1, pp. 69–85.
Pritchard, M. P. & Funk, D. (2010), “The formation and effect of attitude importance in
professional sport”, European Journal of Marketing, Vol. 44 No. 7/8, pp. 1017–1036.
Rice, J. A. (2015), “A generational cohort comparison of endurance sport participant
involvement”, Journal of Applied Sport Management, Vol. 7 No. 3, pp. 47–67.
Rice, J. A., Hambrick, M. E., & Aicher, T. (2018), “A qualitative investigation of sport
activity participation and constraint negotiation among African American endurance
runners”, Journal of Sport Behavior, Vol. 41 No. 1, pp. 1–24.
Ridinger, L. L., Funk, D. C., Jordan, J. S., & Kaplanidou, K. (2012), “Marathons for the
masses: Exploring the role of negotiation-efficacy and involvement on running commit-
ment”, Journal of Leisure Research, Vol. 44 No. 2, pp. 155–178.
Ross, S. (2006), “A conceptual framework for understanding spectator-based brand
equity”, Journal of Sport Management, Vol. 20 No.1, pp. 22–38.
Ross, S. (2008), “Assessing the use of brand personality scale in team sport”, International
Journal of Sport Management & Marketing, Vol. 3 No. 1/2, pp. 23–38.
Ruiz, S. (2019,), “Disney Princess Half Marathon extends sisters’ connection, brings
them closer to late father”, Orlando Sentinel, available at https://www.orlandosentinel.
com/sports/fitness/os-sp-disney-princess-half-marathon-sisters-0223-story.html,
(accessed July 18, 2019).
Running USA (2017), “2017 National Runner Survey”, available at https://www.runningusa.
org/RUSA/News/2018/2018_National_Runner_Survey_Results_Released_by_
Running_USA.aspx, (accessed April 4, 2019).
Running USA. (2016), “2016 State of the Sport: U.S. Race Trends”, available at https://
www.runningusa.org/state-of-sport-us-trends-2015, (accessed April 4, 2019).
Running USA. (2018), “U.S. road race participation numbers hold steady for 2017”,
available at https://www.runningusa.org/RUSA/Research/Recent_Reports/Road_Race_
Trends/RUSA/News/2018/U.S._Road_Race_Participation_Numbers_Hold_Steady_
for_2017.aspx, (accessed April 4, 2019).
Scanlan, T. K., Russell, D. G., Beals, K. P., & Scanlan, L. A. (2003), “Project on elite
athlete commitment (PEAK): II. A direct test and expansion of the Sport Commitment
Model with elite amateur sportsmen”, Journal of Sport & Exercise Psychology, Vol. 25
No. 3, pp. 377–401.
Scanlan, T. K., Russell, D. G., Scanlan, L. A., Klunchoo, T. J., & Chow, G. M. (2013),
“Project on elite athlete commitment (PEAK): IV. Identification of new candidate
commitment sources in the Sport Commitment Model.” Journal of Sport & Exercise
Psychology, Vol. 35 No. 5, pp. 525–535.
Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J., Chon, K., Claiborn, C. B., Johar,
J. S., & Berkman, H. (1997), “Assessing the predictive validity of two methods of
measuring self-image congruence”, Academy of Marketing Science Journal, Vol. 25
No. 3, pp. 229–241.
 Congruity Perceptions 217

Sirgy, M. J., & Su, C. (2000), “Destination image, self-congruity, and travel behavior:
Toward an integrative model.” Journal of Travel Research, Vol. 38 No. 4, pp. 340–352.
Summers, J., Machin, V., & Sargent, G. (1983), “Psychosocial factors related to marathon
running”, Journal of Sport Psychology, Vol. 5, pp. 314–331.
Summers, J. J., Sargent, G. I., Levey, A. J., & Murray, K. D. (1982), “Middle-aged, non-elite
marathon runners: A profile”, Perceptual and Motor Skills, Vol. 54 No. 3, pp. 963–969.
Tajfel, H., & Turner, J. C. (2010), “An integrative theory of intergroup conflict”, In T.
Postmes, N. R. Branscombe, T. Postmes, & N. R. Branscombe (Eds.), Rediscovering social
identity, Psychology Press, New York, NY, pp. 173–190.
Chapter 12

Service Provider Selection with

the Analytic Hierarchy Process
An Approach to Sport Facility
Enterprization in China
Yunxia Ding and Yelena Y. Wang

Introduction
Public sport facilities, as the material foundation of national fitness, play an impor-
tant role in the development of Chinese sport industry. In order to overcome the
business difficulties that may follow mega events, such as the 2008 Beijing Olympics,
the 2010 Guangzhou Asian Games, and National Games over the years, opera-
tional reforms are necessary. Facing the coming challenges such as the 2022 Beijing
Winter Olympics and the 2022 Asian Games at Hangzhou, China is hastening to
streamline the administration of the facilities to ease the contradiction between
inefficient supply and unsatisfied demand of public sport services.
In May of 2016, the General Administration of Sport of China released The
13th Five-Year Plan of Sport Development and brought forward important strat-
egies for introducing and applying the modern enterprise system, regenerating
public stadiums, exploring the separation of ownership and operation of major
stadiums, and promoting reforms of public stadium management and operation
mechanisms (GASC, May 5, 2016). Soon afterward, the 13th Five-Year Plan of
the Sport Industry proposed maintaining state-ownership of sport facilities while
transferring operating rights (GASC, July 13, 2016). This transformation was
again emphasized in The Outline of the Healthy China 2030 Plan (CPC central
committee & Chinas State Council, October 25, 2016). Subsequently, the GASC
formulated The Work Programme on Further Reinvigorating Stadium Resources and
started trials in Jiangsu and Zhejiang provinces and the Chongqing municipal-
ity. These trials focused on sport facility management reforms, maintaining pub-
lic stadiums as state-owned while granting operating rights to private entities.
Through the innovation of institutional superstructure design and field explora-
tion, the Chinese sport industry decision makers sought to advance the develop-
ment of public stadiums, improve the capacity of sport services that they provide,
pursue a new goal of fitness-for-all, and accelerate the development of the fitness
and recreational sports industry.
According to the law of the market economy, any decision-making depends on a
comparison of costs and benefits. The key to social investment and financing is the
reasonable expected return on investment (ROI) (Ye, 2008). Public sport facilities

DOI: 10.4324/9781003302476-12
 Service Provider Selection 219

are usually large- and medium-scaled, which require large investments. The related
projects are long term in most cases. Therefore, only ROI above the average profit
margin can attract private actors’ interest. In order to address developing prob-
lems, such as financial losses in operating and maintaining local sport stadiums
and arenas after mega games, specific business pattern reforms are necessary for
the Chinese sport industry. As Xue and Mason (2019) said that the Chinese tradi-
tional sport facility operating pattern is different from that of their peers in North
American cities, very few sport arenas in Chinese cities host home teams. After
mega games, these facilities’ expected ROI from day-to-day operations faces more
realistic challenges. Newly developing multi-purpose sports complexes are emerg-
ing in response to the proper conditions. Revenues from multi supplementary busi-
nesses are expected instead of simply from sport events. Also, more benefits for the
community to improve residents’ quality of life need to be implemented.
High hopes are placed on the development of sports complexes. In this business
mode, a general contractor (i.e., the service integrator) invites a group of subcon-
tractors (i.e., sub-service providers) to build a service supply chain together. All
service providers (e.g., sports training companies, entertainment corporations,
and catering enterprises) contribute to the entire facility’s operational effective-
ness. All providers’ service capabilities, standardization levels, and qualities are
important. However, every subcontractor is usually single-functional in a certain
space. Thus, resource integration is necessary to establish a service ecosystem,
which improves the operational efficiency of public sport facilities and meets the
multi-level and diverse sports service demands of consumers.
The selection of subcontractors is important in order to increase the entire flex-
ibility of sport facilities and improve competitive power. How to make an effective
selection of service providers within the supply chain in the sports industry also
captures researchers’ attention. In this study, the authors established a service
provider selection model with an analytic hierarchy process. This study is the first
AHP research to build a reliable and valid scale for service provider selection in
terms of large- and middle-sized sport facilities in China. Through this model,
service integrators are able to evaluate their potential subcontractors. To our
knowledge, this model is the first applicable and sports-focused service provider
selection model regarding the special market of China, which is expediting the
pace of the supply-side reforms. The authors hope that our findings can help the
Chinese sport industry to take a positive step in its enterprization process.

Review of Literature

The Supply Chain

The concept of the supply chain and supply chain management was consid-
ered the first important aspects within the theoretical framework of this study.
The concept itself is not new. Supply chain was proposed in the early 1980s
220 Y. Ding and Y. Y. Wang

(Oliver & Webber, 1982) and has become a wide research focus since the late
1990s (e.g., Beamon, 1999; Ellram et al., 2004; Fisher, 1997; Mentzer et al.,
2001). Numerous definitions have been provided.
Handfield (1999) stated that the supply chain involves “all activities asso-
ciated with the flow and transformation of goods from the raw materials stage
(extraction), through to the end user, as well as the associated information flows.
Material and information flow both up and down the supply chain”, and “supply
chain management (SCM) is the integration of these activities through improved
supply chain relationships, to achieve a sustainable competitive advantage”
(p. 2). Simchi-Levi et al. (2008) stated generally that SCM “is a set of approaches
utilized to efficiently integrate suppliers, manufacturers, warehouses, and stores so
that merchandise is produced and distributed at the right quantities, to the right
locations, and at the right time, in order to minimize system-wide costs while sat-
isfying service level requirements” (p. 1). SCM also refers to the logistics network
(p. 1). Based on existing studies, the authors adopted the SCM approach, which
builds competitive strategic alliances. Fundamentally speaking, the purpose of
SCM is to satisfy the diverse demands of consumers and, therefore, achieve over-
all competitiveness.
Some Chinese researchers have noted the rationality of Western economic
theories and began to introduce the market-based operation pattern to the
Eastern world. Since the Chinese market leans on traditional industries, most
studies with regard to supply chain and SCM have focused on manufacture (Yang,
2006), agriculture (Zhu, 2004), and the real estate industry (Kang, 2007). Only a
limited number of studies with regard to the service supply chain can be found in
China (Yu, 2011), while many Western scholars have paid attention to compar-
ing the differences between manufacturing and service supply chain performance
(Sengupta et al., 2006), performing case studies of the after-sales service supply
chain (Saccani et al., 2007), and examining the service supply chain manage-
ment (Ellram et al., 2004; Stentoft Arlbjørn et al., 2011). Only a few pioneers
can be found in the Asian context recently, such as Cho et al. (2012) and Wang
et al. (2015). In China, even fewer scholars have started some preliminary explo-
rations (Huang et al., 2013; Yan et al., 2005). A significant research gap still exists.
Especially, the sport industry, an important component of the service industry,
has not been well considered yet, to our knowledge.

Supplier Selection and Assessment

Supplier selection has been a widely studied research system all over the world
for more than half a century. For example, Dickson (1966) studied vendor selec-
tion systems that existed at that time, analyzed more than 50 features of vendor
performance that had been presented in a variety of purchasing literature and
proposed a 23-factor criterion in which quality, delivery, and performance his-
tory were the top three weighted indexes. In Johnson, Meade, and Rogers’ study
(Liu & Zhang, 2011), time, quality, cost, and service were deemed as the success
 Service Provider Selection 221

factors in partner selection. Moreover, Kannan and Tan (2002) divided their cri-
terion into two aspects: soft and hard. Their empirical study found that the soft
aspect (i.e., non-quantifiable factors, such as a supplier’s strategic commitment
to a buyer) plays a more significant role than the hard aspect (i.e., quantifiable
factors, such as supplier capability).
Most existing studies have focused on product suppliers’ selection, and var-
ious assessment criteria have been proposed, while only a few were applied
to the service market (e.g., Billhardt et al., 2007; Ko & Pastore, 2005). Since
more business strategic partnerships are being established between product
and service enterprises, suppliers and their provisions have become increas-
ingly specific. Different supplier selection criteria must be developed in dif-
ferent domains, being more applicable. For instance, Kan and Chen (1998)
suggested that an auto parts supplier assessment criterion should consider
quality, delivery, flexible manufacturing, the balance between delivery time,
price, and quantity, and diversity. Zeng et al. (2007) provided a four-dimensional
transport service supplier selection model that involved quality, service, price,
and delivery time. Moreover, Li and Wang (2015) established a criterion for
the retailer-led supply chain that involved cost, flexibility, quality, technology,
risk, and environment.
In the Chinese sport context, only Wang et al. (2013) have analyzed the
demand of mega events, discussed the goal of establishing a supply chain that
focuses on mega sport events and mentioned the process of selecting and
assessing service providers. The importance of establishing criteria for service
provider selection has been emphasized by Wang et al. (2013); however, they
have not proposed any criteria yet.. On the one hand, several well-known
enterprises, such as Amway, General Motors, Siemens, Henkel, Philips, and
Sumitomo Mitsui, explained from the perspective of being global suppliers at
the 2003 International Sourcing Summit Forum that their criteria involved
seven major assessment standards: assurance of quality, speed of delivery,
reasonable price, communication capabilities of foreign trade, ECO certifi-
cation, application of E-commerce, and secure quality of service (He, 2003).
On the other hand, from the perspective of selecting suppliers, Carrefour and
Walmart stated that their standards included an examination of whether
potential suppliers have export authority, price competitive advantage, qual-
ity assurance, mass productivity, a spirit of continuous learning, and on-time
delivery (Chen & Qi, 2009).
According to all of the above studies, no accepted general standard of supplier
selection can be determined. A specific set of criteria must be applied in a spe-
cific realm. Nevertheless, no literature with regard to the sport service provider
selection, to our knowledge, has been done yet in Western or Eastern academia.
Therefore, this study is the first research to propose an applicable service provider
selection model. This study proceeds from the perspective of establishing a sport
service supply chain and spotlights enterprization reforms in the Chinese sport
industry.
222 Y. Ding and Y. Y. Wang

The Stakeholder Theory

The stakeholder theory also guided this study as the third important aspect
of the theoretical framework. Since the landmark book Strategic management:
a stakeholder approach by Freeman was published in the early 1980s, the stake-
holder theory has become a theoretical basis of various studies (e.g., Donaldson &
Preston, 1995; Jensen, 2010). Freeman (1984) suggested that a company should be
concerned about all of its stakeholders as he wrote the original concept that was
stated by the Stanford Research Institute (1963), in which stakeholders are “those
groups without whose support the organization would cease to exist.” (p. 31).
Then, he proposed more formally that a stakeholder can be “any group or individ-
ual who can affect or is affected by the achievement of the organization’s objec-
tives” (p. 46). In other words, these groups both directly and indirectly contribute
to the operation of an enterprise and encompass customers, employees, suppliers,
political action groups, environmental groups, local communities, media outlets,
financial institutions, governmental groups, and more. Stakeholders affect the
implementation of the enterprise objective and the operation’s performance;
in the meantime, the stakeholders’ own interests are affected by the latter two
groups (Freeman, 1984). Hendry (2001) emphasized that an organization cannot
develop properly without the contributions of all stakeholders. Similarly, Blair
and Stout (1999) suggested that each stakeholder invests from the perspectives of
fairness, reasonableness, and balance. Stakeholders take responsibilities together
and share the entire profits. In this way, a long-term effective benefit distribution
mechanism can be established (Ensign, 2001).

Purpose of Study
Due to the complexity of SCM of large- and middle-sized sport facilities and
the diversity of all of their participants, the authors adopted both the concept
of supply chain and the stakeholder theory to analyze the stakeholders’ benefit
appeals and to examine how the ecosystem of sport facilities can be optimized.
Therefore, initially, we need to demarcate two groups of stakeholders, including
service integrators and service providers. According to the concept of SCM and
the stakeholder theory, a key stakeholder must exist in each supply chain as the
center of interest subjects. This key stakeholder coordinates all other subjects in
the chain in order to harmonize the strategic partnership. The key stakeholder
must also understand how to utilize the internal and external resources and help
all subjects in the supply chain to achieve a win–win situation. In the domain of
sport services, the service integrator who has operation rights plays the role of the
key stakeholder.
Through fixed investment, the functional layout is planned, and certain facil-
ities are arranged and installed in the sports complex by its contracted integra-
tor. Then, the integrator invites potential sub-service suppliers – both sports and
related supporting services – to join the chain. In this way, a supply-chain-based
 Service Provider Selection 223

sports complex mode can be run. The service integrator (i.e., the general contrac-
tor) as the key stakeholder establishes the service ecosystem by uniting all service
providers. Therefore, the utilization and profitability of the sport facilities and the
general competitive power of the complex are closely related to the integrator’s
performance. Strong capability of integration and standard management skills
are necessary.
In an effort to separate operation rights from ownerships, privatization reforms
are promoted to address the issues of Chinese traditional state-centered economic
mechanism of public sports facilities. In order to develop each sports complex
with its own features and optimize the service ecosystem, every single element
of the supply chain needs the cooperation of different professional market sub-
jects. These subjects are diverse service providers within the complex. They build
the foundation of the chain. The service capability, standardization level, and
quality of service of every single element impact the performance of the entire
service chain. For example, hosting a Chinese Super League soccer tournament
in Hongkou Football Stadium – a typical sports complex in Shanghai – requires
that dozens of sports entities (e.g., the operator of the stadium, professional soccer
clubs, the operator of the event, broadcasters, technology suppliers, fan service
providers, sponsors) cooperate. The majority of these market entities is based in
Hongkou Football Stadium or are located nearby. These entities have developed
a cooperative effort as a community of interests.

Method
A case study, which is a qualitative research method, is based on a preliminary
analysis of the objects. Researchers choose typical cases for in-depth investigation
instead of a large sampling survey to explore new questions or new circumstances.
In this study, the authors conducted a typical survey through field visits in ten
sports complexes in six cities, in three provinces, including Jiangsu Wutaishan
Sports Center, Nanjing Olympics Sports Center, Jiangsu Hongshan Sports Park,
Shanghai East Asia Sports Center, Shanghai Hongkou Soccer Stadium, Shanghai
Xiang Cube Multi-Purpose Sports Complex, Zhejiang Huanglong Sports Center,
Zhejiang Hangzhou Gymnasium, Zhejiang Taizhou Sports Center, and Zhejiang
Shaoxing Olympic Sports Center. The sports complexes that we chose are the
typical successful cases in China. Many of them are on the recommendation list
provided by the GASC (2020).

Questionnaire
We primarily used a quantitative method to analyze the collected data. Before
that process, a pre-qualitative analysis (an empirical assessment of expertise) was
utilized to determine the initial scale items. The quantitative analysis was con-
ducted through SPSS to obtain descriptive statistics results and to test the reli-
ability and validity of the study. Corrected item-total correlations and principal
224 Y. Ding and Y. Y. Wang

component analysis were applied to the refining of the indexes of the proposed
model.

Analytic Hierarchical Process

One of the main structured models is analytic hierarchical process (AHP), which
comes from mathematics and psychology and was originally proposed by Thomas
L. Saaty in the 1970s and was widely applied to support the decision-making the-
ory in various domains for decades (e.g., Guariguata et al., 2014; Handfield et al.,
2002). AHP helps researchers to develop scales through pairwise comparisons
(Saaty, 2008). Saaty explained that these scales measure intangibles in relative
terms. The comparisons are made by using a scale of absolute judgments that
represent the extent to which one element dominates another with respect to
a given attribute. In this study, following the review of literature, interviews of
experts, and statistical analyses, the authors proposed a preliminary service pro-
vider selection model. Then, we examined this model in terms of AHP and gave
different weights to different evaluation indexes in order to determine the status
of each index in the entire index system. Therefore, the degree of each index’s
impact can be discovered.

Outline of the Initial Decision Hierarchy

Although various supplier selection studies have been conducted that focused
on enterprise procurement, no extant literature focused on sport facilities. Initial
items were adapted from existing literature, such as Chen et al. (2012), Saaty
(2008), and Song and Huang (2010). Because of many recent development pro-
grams, such as sports towns, outdoor sports bases, sports and leisure bases, and
sports tourism complexes, we structured our decision hierarchy from the top with
the goal of selection of service providers, which is followed by objectives from
three broad perspectives: contract fulfillment (from the perspective of the service
provider), cooperation within the supply chain (from the perspective of the supply
chain), and consumer demand satisfaction (from the perspective of the consumer),
through the intermediate level (seven criteria on which subsequent elements
depend) to the lowest level (a set of 41 alternatives). Most of the items of contract
fulfillment and consumer demand satisfaction were adopted from the existing liter-
ature while the items of cooperation within the supply chain were selected by the
authors according to the empirical assessment of expertise. Details are provided
in Table 12.1.

Refinement of the Initial Decision Hierarchy

After the proposal of an initial hierarchy of sport service provider selection
(7 intermediate criteria, 41 alternatives), which was adapted from the existing
literature, we created a survey for a pretest to refine the index system. All
Table 12.1 The Initial Decision Hierarchy of Sport Facilities Service Provider Selection

Objectives from Qs (Is It Important Whether

Goal Broad Perspectives F Lowest Level Alternatives the Service Provider Has/Can/Is…?)

Sport facilities, Contract Brand (A11) Industry Influence (A111) (Dickson, 1966) Influence and a good brand position
service provider fulfillment (from in the particular industry
selection (A) the perspective Service quality (A112) (Song & Huang, 2010) High quality of its service product
of the service Public Image (A113) (Ding et al., 2021) Social responsibility and public
provider) (A1) service achievements
Cooperate culture (A114) (Roy et al., 2004) A good company philosophy
Entrepreneurship (A115) (Wasti & Liker, 1997) Some leaders being proved successful
Level of credibility (A116) (Li, 2006) Credit record
Policy support (A117) (Xu & Han, 2015) Support from the government
Management Management team (A121) (Wagner & An efficient management team
(A12) Hoegl, 2006)
Professorial level (A122) (Dickson, 1966) Professorial management
Organization (A123) (Wang & Wang, 2005) A reasonable and friendly internal
relationship
Salary system (A124) (Huang,Yi, & Jie, 2016) A reasonable salary system

Service Provider Selection

Innovation (A125) (Chen & Qi, 2009) Some valuable proprietary marks or
intellectual property rights
Cross-industry collaboration (A126) (Hartley Experiences working with peer
et al., 1997) companies from other industries
Marketing team (A127) (“Guidelines on A good marketing team
promoting the development of region-wide
Tourism”, 2018)
Integration Capability (A128) (Song & Huang, 2010) The capability of resources integration
Flexibility (A129) (Hu, 2005) Capability to deal with different
situations in management

225
(Continued)
 226 Y. Ding and Y. Y. Wang
Table 12.1 The Initial Decision Hierarchy of Sport Facilities Service Provider Selection (Continued)

Objectives from Qs (Is It Important Whether

Goal Broad Perspectives F Lowest Level Alternatives the Service Provider Has/Can/Is…?)

Performance Asset (A131) (Dickson, 1966) Enough net assets and a good related
(A13) financial position
Debt (A132) (Xiao & Liao, 2010) A reasonable asset-liability ratio
Flows (A133) (Dou et al., 2016) Reasonable inflow and outflow of
corporate cash to carry out
operations, investment, and
financing
Dynamic Performance (A134) (Wasti & A good operating income and profit
Liker, 1997) in the past three years
Cooperation Experience (A135) (Liu et al., Previous positive cases of
2003) cooperation and relatively low
negative effect
Industry Trends (A136) A positive industry trend
(Wynstra, & Ten Pierick, 2000)
 Service Provider Selection 227

41 alternatives were measured on a ten-point Likert-type scale. The participants

included owners of the sport complexes, operation directors, decision makers of
relevant government departments, decision-making consultants, and academic
experts in the field of sports management. All selection indicators were described
specifically. Also, the authors provided further explanations when needed during
the questionnaire distribution and collection. Fifty questionnaires were distrib-
uted, and 48 were received. 14 of the responses were obtained by the researchers
in the field, and 34 were collected online. Most participants were male (94%).
All of them had college degrees, seven with master’s degrees, four had PhDs, and
two were doctoral candidates. Most participants had relatively long employment
experience in their specific positions (≤5 years: 1, 5–10 years: 18, 10–20 years: 20,
and ≥20 years: 9). Among them, six pieces with obvious bias (e.g., uncompleted
or questionnaires which showed answers with obvious regularity) were removed.
Therefore, a total of 42 valid samples were remained with the 84% effective rate.
Then, statistical analyses were conducted through SPSS. Corrected item-total
correlations and principal component analysis were utilized to refine the initial
hierarchical model.
During the data collection, some experts pointed out that the concepts of sev-
eral indicators overlapped, while some classifications were disputable. As they
suggested, we narrowed down the model, deleted the seven intermediate level
criteria, and directly linked the lowest level items to the three objectives from
broad perspectives, such as sub-criteria and criteria. Moreover, we refined the
model according to the rule of thumb, which holds that if we found an item
with a corrected item-total correlation (CITC) value below .50 and a Cronbach’s
α below .70, we removed this item if there was no existing evidence from the
literature that strongly supported the item (Churchill, 1979). SPSS 19.0 was used
to examine the CITC and Cronbach’s α values of the 41 items under each of the
seven factors.

Results
According to the rule of thumb described above, four items (A112, A114, A115,
A117) under the factor of Brand (A11) were removed. Under the factor of man-
agement (A12), two items (A128, A129) were removed. Three items (A131, A133,
A134) were removed from the factor of performance (A13), while all four items
under the factor of matching (A21) were retained. Under the factor of value
enhancement (A22), two items (A221, A225), which CITC values .374 and .476,
were removed. The items under the factor of market response (A31) were all
retained. And three items (A323, A324, and A325) under the last factor, technol-
ogy innovation (A32), were removed to meet the statistical requirements. After
the reduction, all 27 retained items under the seven factors have CITC values
over .50. And the Cronbach’s α for each factor was over .70. The internal con-
sistency reliability of the model after item deleted is acceptable, which meets the
requirements of the rule of thumb (Table 12.2).
 228
Table 12.2 CITC and Reliability Analysis of All 7 Intermediate Level and 41 Lowest Level Criteria

Index Code Corrected Item-Total Cronbach’s α Factor α CITC if Item Cronbach’s α if Factor α if Item

Y. Ding and Y. Y. Wang

Correlation Deleted Item Deleted Deleted

Brand (A11) A111 0.619 0.604 0.685 0.573 0.659 0.741

A112 0.327 0.670
A113 0.541 0.616 0.542 0.683
A114 0.464 0.632
A115 0.283 0.676
A116 0.515 0.614 0.613 0.613
A117 0.184 0.732
Management (A12) A121 0.691 0.839 0.862 0.705 0.825 0.855
A122 0.687 0.839 0.738 0.819
A123 0.611 0.845 0.579 0.841
A124 0.618 0.845 0.647 0.831
A125 0.631 0.844 0.587 0.841
A126 0.511 0.855 0.472 0.856
A127 0.649 0.841 0.642 0.832
A128 0.442 0.860
A129 0.493 0.857
Performance (A13) A131 0.584 0.658 0.731 0.613 0.732 0.789
A132 0.462 0.695
A133 0.660 0.633 0.743 0.584
A134 0.526 0.678 0.544 0.800
A135 0.187 0.763
A136 0.404 0.716
Matching (A21) A211 0.719 0.669 0.798 0.719 0.669 0.798
A212 0.709 0.713 0.709 0.713
A213 0.727 0.650 0.727 0.650
A214 0.824 0.446 0.824 0.446
(Continued)
Table 12.2 CITC and Reliability Analysis of All 7 Intermediate Level and 41 Lowest Level Criteria (Continued)

Index Code Corrected Item-Total Cronbach’s α Factor α CITC if Item Cronbach’s α if Factor α if Item
Correlation Deleted Item Deleted Deleted

Value enhancement (A22) A221 0.374 0.781 0.772 0.789

A222 0.603 0.712 0.587 0.785
A223 0.705 0.673 0.766 0.595
A224 0.585 0.716 0.600 0.785
A225 0.476 0.756
Market response (A31) A311 0.830 0.793 0.872 0.830 0.793 0.872
A312 0.847 0.784 0.847 0.784
A313 0.649 0.866 0.649 0.866
A314 0.592 0.885 0.592 0.885
Technology innovation (A32) A321 0.699 0.690 0.768 0.582 0.660 0.746
A322 0.595 0.711 0.585 0.656

Service Provider Selection

A323 0.329 0.783
A324 0.394 0.761
A325 0.480 0.743
A326 0.622 0.705 0.565 0.671

229
230 Y. Ding and Y. Y. Wang

Because this study has a limited sample size, a higher reliability is required.
Thus, we re-tested the reliability of the model after the first round of screening to
get the CITC and Cronbach’s α values of the 27 remaining items. Also, we com-
pared the Cronbach’s α values of the overall model before and after the second
round of screening. Therefore, 12 more items (A122, A123, A124, A125, A126,
A127, A133, A211, A212, A222, A313, and A326) were removed because their
CITC values were below .50. After the double refinement, the total Cronbach’s α
increased from .923 to 0.930. As a result, all retained 15 items’ CITC values met
the requirements of the rule of thumb (as shown in Table 12.3).

Index Weight Value Analysis

The index weight of an item quantificationally indicates the importance of this
item in the measuring model. How to determine the index weight is the core

Table 12.3 C
 ITC and Reliability Re-Analyses of the 27-Item Model after the First
Reduction

Variable CITC Cronbach’s α Total α CITC If Item Deleted Total α If Item Deleted

A111 0.552 0.921 0.923 0.511 0.930

A113 0.613 0.919 0.647
A116 0.617 0.919 0.591
A121 0.608 0.920 0.641
A122 0.486 0.921
A123 0.345 0.922
A124 0.311 0.924
A125 0.461 0.923
A126 0.455 0.922
A127 0.479 0.920
A131 0.589 0.920 0.616
A133 0.414 0.921
A134 0.528 0.921 0.508
A211 0.445 0.922
A212 0.486 0.921
A213 0.565 0.922 0.597
A214 0.646 0.919 0.670
A222 0.417 0.923
A223 0.564 0.920 0.502
A224 0.518 0.921 0.508
A311 0.700 0.918 0.730
A312 0.750 0.917 0.747
A313 0.469 0.917
A314 0.753 0.918 0.755
A321 0.659 0.919 0.711
A322 0.565 0.920 0.591
A326 0.361 0.920
 Service Provider Selection 231

of this study. The authors followed Saaty’s AHP pairwise in comparing each
item. The index weights and an overall ranking of the 15 items were determined
through a fuzzy comprehensive evaluation method that tested the proposed hier-
archical index system. According to the model refinement and accuracy con-
cerns that were identified by the experts who have participated in the pretest, the
first-order sub-criteria (i.e., the intermediate level criteria in the initial model of
Table 12.1) have been removed from further analysis. The specific four steps are
as follows.

First, Structure the Decision Hierarchy

Based on the reduced model with items deleted, an analytic hierarchy sport facil-
ities service provider selection model from the top with the goal of the decision,
through the three criteria, to the 15 sub-criteria was developed (Table 12.4). The
three criteria involve the three dimensions of structure factors: Contract ful-
fillment (i.e., provider-based), coordination (i.e., supply chain-based), consumer
demand satisfaction (i.e., consumer-based). Also, four to six sub-criteria reside
under each structure factor. This design meets the suggestion of recent studies
that kept the number of factors as small as possible. Therefore, the survey will
not bother respondents too much with a long list of comparisons (Lee & Ross,
2012). As Saaty and Ozdemir (2003) suggested, this analytic hierarchy model can
be considered to be efficient since the number of elements in each group was no
more than seven.

Table 12.4 T
 he Initial Analytic Hierarchy Model of Sport Facilities Service Provider
Selection

Purpose Criteria Sub-Criteria

Sport facilities service Contract fulfillment Industry influence (U11)
provider selection (U) (provider-based factors) Public Image (U12)
(U1) Level of credibility (U13)
Management team (U14)
Asset (U15)
Dynamic performance (U16)
Coordination (supply Target market (U21)
chain-based factors) (U2) Operation pattern (U22)
Industry driving (U23)
Consumption stimulation (U24)
Consumer demand Adaptive capacity (U31)
satisfaction (consumer- Content innovation (U32)
based factors) (U3) Professionals (U33)
Standards (U34)
Customization (U35)
232 Y. Ding and Y. Y. Wang

Second, Construct a Set of Pairwise Comparison Matrices

For comparisons, Saaty’s (1980) 1–9 fundamental scale was used, as shown in
Table 12.5. Mathematically, the algebraic matrix of paired comparisons is depicted
by the following expression:

 w1 /w1 w1 /w2  w1 /w n 
 
 w2 /w1 w2 /w2  w2 /w n 
A= 
    
 w n /w1 w n /w2  w n /w n 
 

where wi /wj is the relative importance for the element i to the element j. In other
words, wi and wj are the contributions of element i and the element j on the
objective U.
Considering both the expertise requirements of this study and the research
limitations into consideration, we recruited ten experts in the field, which
included three scholars from the leading sports management programs in China’s
universities, two department heads of provincial sports bureaus, and the directors
of operations of five large-scale sport complexes. In order to calculate w for the
matrix, the judgments of each element made by the experts were averaged, and

Table 12.5 The Fundamental Scale of Absolute Numbers

Intensity of Importance Definition Explanation

1 Equal importance Two elements contribute equally to

the objective
3 Moderate importance Experience and judgment slightly
weight one element over another
5 Strong importance Experience and judgment strongly
weight one element over another
7 Very strong importance An element is weighted very strongly
over another; its dominance
demonstrated in practice
9 Extreme importance The evidence weighting one element
over another is of the highest
possible order of affirmation
2, 4, 6, 8 Intermediate values between the two
adjacent judgments
Reciprocals of above If activity i has one of the above
non-zero numbers assigned to it
when compared with activity j, then
j has the reciprocal value when
compared with i
 Service Provider Selection 233

the relative weight results were rounded to two decimal places. The 3 by 3 recip-
rocal matrix of the first-order criteria is depicted as:

 1.00 1.55 2.20 

 
A =  0.65 1.00 1.70 
 0.45 0.59 1.00 
 

And the specific matrices of sub-criteria under each structural factor Ai are
depicted as:

 2.67 0.49 0.75 1.00 0.35 0.48 

 
 1.00 0.58 0.66 0.37 0.32 0.87 
 3.11 2.50 2.53 2.82 1.00 3.26 
A1 =  
 1.72 1.00 2.20 2.03 0.40 2.57 
 1.51 0.46 1.00 1.34 0.40 1.35 
 1.15 0.39 0.74 2.07 0.31 1.00 
 

 0.86 1.00 1.24 0.47 0.65 

 1.00 0.90 0.55 0.32   
   1.79 2.12 2.17 1.00 2.15 
1.11 1.00 1.00 1.00
A2 =   A3  1.15 1.55 0.81 0.47 1.00 
 1.82 1.00 1.00 0.98   
   1.00 1.17 0.98 0.56 0.87 
 3.16 1.00 1.03 1.00   1.02 0.81 1.00 0.46 1.24 

Third, Consistency Analysis of Interval Judgment Matrix

Since decisions that are made by humans are likely inconsistent, how to measure
and improve the consistency of judgments is emphasized by AHP researchers
(Lee & Ross, 2012; Saaty, 2008). Several measures, such as Consistency Index
(CI), Random Consistency Index (RI), and Consistency Ratio (CR) that were
provided by Saaty have been widely accepted (Lee & Ross, 2012; Teknomo,
2006). Specifically, the largest Eigen value is equal to the size of comparison
matrix, λmax = n, and CI = (λmax − n)/(n − 1). Then, a researcher can use the CI to
compare with the appropriate index which is called the Random Consistency
Index (RI) to determine whether the Consistency Ratio (CR) is above 10% or
less. The comparison can be depicted in formula CR = CI/RI (Saaty, 1980,
1982). If the CR is found to be more than 10%, Saaty (1980, 1982) stated
that the judgments might need to be revised. Only when the CR is ≤10% is
a good consistency indicated, and the weight vectors are the eigenvectors.
Our results of the average RI (n = 1–10), respectively, equals 0, 0, 0.58, 0.90,
1.12, 1.24, 1.32, 1.41, 1.45, and 1.49. The Consistency Ratio results are given
in Table 12.6.
234 Y. Ding and Y. Y. Wang

Table 12.6 The Consistency Ratio Results Matrices

Pair-Wise Comparison Matrices Consistency Analysis

Matrix

Index A A1 A2 A3

Λmax 3.00 6.29 4.10 5.06

CI 0 0.06 0.03 0.01
RI 0 1.26 0.89 1.12
CR 0.00 0.05 0.04 0.01
Assessment yes Yes yes Yes

According to the results, we noticed that Matrix A showed perfect consistency

with a CR of .00, while Matrix A1, A2, and A3 had CR, respectively, equal to
.05, .04, and .01. Consistency was indicated among the structural matrix and all
measurement matrices.

Forth, Determine the Index Weight

The process of decision-making in the selection of sport service providers requires
us to determine the index weight. Through yaahp, a software for AHP, we found
weights for all of our indexes (Figure 12.1).

Fifth, Index Weight Assessment, Item Grading, and Reranking

According to the results we found, which were shown in Figure 12.1, the authors
re-ranked the measurement indexes based on their own weights and calculated
the cumulative weights. A Pareto analysis of variance was employed, and the
measurement indicators were graded for their respective impact. The Pareto

Figure 12.1 AHP Index System of Sport Facilities Service Provider Selection.
 Service Provider Selection 235

Table 12.7 W
 eight Assessment and Item Grading of the Refined Sport Facilities Service
Provider Selection Index System

Measurement Cumulative
Order Index Own Weight Weight Grade Parent Structure Index

1 Level of credibility 0.1629 0.1629 A Contract fulfillment

2 Consumption 0.1058 0.2687 A Coordination within
stimulation the supply chain
3 Management team 0.0986 0.3673 A Contract fulfillment
4 Industry driving 0.0883 0.4556 A Coordination within
the supply chain
5 Target market 0.0804 0.536 A Coordination within
the supply chain
6 Content innovation 0.0681 0.6041 A Consumer demand
satisfaction
7 Asset 0.0604 0.6645 A Contract fulfillment
8 Dynamic performance 0.0555 0.72 B Contract fulfillment
9 Industry influence 0.0545 0.7745 B Contract fulfillment
10 Operation pattern 0.0498 0.8243 B Coordination within
the supply chain
11 Public image 0.0413 0.8656 B Contract fulfillment
12 Standardization 0.0359 0.9015 C Consumer demand
satisfaction
13 Customization 0.034 0.9355 C Consumer demand
satisfaction
14 Professionals 0.0335 0.969 C Consumer demand
satisfaction
15 Adaptive capacity 0.031 1 C Consumer demand
satisfaction

Principle (i.e., the “80/20 Rule”) said that 80% of a project’s benefit comes from
key contributions, which account for only 20% of all work (Kaliszewski, 2012).
Therefore, we sorted the 15 sub-criteria into three grades in descending order.
The indexes with a cumulative frequency of 0%–70% are Grade A indicators,
and the indexes within the range of 70%–90% are secondary indicators
(Grade B), and the indexes from 90% to 100% are Grade C indicators. In gen-
eral, seven items (level of credibility, consumption stimulation, management
team, industry driving, target market, content innovation, and asset) were deter-
mined as the key indicators, and the remaining eight variables were general
indicators (Table 12.7).

Discussion
When sport facility integrators apply the proposed service provider selection
model through AHP to their future assessment and decision-making, they may
follow up with some corresponding adjustments based on the differences in the
focuses of their specific sports complex. Subsequent performance evaluation is
236 Y. Ding and Y. Y. Wang

needed. Indeed, the selection of service providers is a multi-goal, multi-criteria

decision, especially when the enterprization of sport facilities in China is still
in its early stages. A time-proven application is necessary. Industry practitioners
should adjust measures to local conditions and seek the most suitable criteria for
themselves and consider city resource conditions, the proportion of sport and
support services that they plan to offer, the future developing focus on their core
activities, industry attributes, and trends of their key investment projects.
Urban areas will likely continue to be the main investment and development
front regarding most sport facilities in China for a long time. However, not all cit-
ies can be considered similar. There are various population sizes, land resources,
existing sports and related industry features, sports culture bases, traffic condi-
tions, and different markets that the sport facilities target. Sports complexes in
different cities and even in different districts within one city may require differ-
ent developing patterns. A complex should establish a focused service provider
system that accords with the complex’s internal operation eco-mechanism and
the surrounding market and determine service content through comprehensive
consideration. Only in that way can efficient development with local variables be
expected.
According to various carrier types and relevant structures, sport facilities in
China can be divided into two categories: one type of complex in which sport
services play a leading role and the other type in which sport services provide
support. Specifically, a sport-service-leading complex is reformed based on new or
existing large and medium-sized stadiums. In order to develop this type of com-
plex, the original functional properties should not be changed; meanwhile, more
sport service projects providers can be introduced and attract other related indus-
try service providers to join the supply chain to achieve an all-win situation. In
contrast to the above type, in order to develop a complex which uses sports as
supporting services, decision makers should highlight their complex’s commercial
framework, inlaying sport-related services into the existing business services. In
this way, the various extant service content will be enriched, and the commercial
pattern will be optimized. In sport-supporting type of complexes, sport services
are considered to be important because of their external influence. Therefore,
fitness-for-all and commercial benefits can be promoted simultaneously.
From case studies that were done previously by the authors in terms of resource
endowments and developing directions, the following five modes of sport facilities
were developed: (1) event-oriented, (2) fitness and recreation-focused, (3) urban,
comprehensive service support, (4) local sport industry incubator, and (5) sport
tourism-oriented. A service integrator needs to consider both the supply side and
the demand side of a sports complex’s service content in order to optimize the
industrial ecology and build an industrial ecosystem within the facility.
Sport activities can be generalized into two categories: daily workout and fash-
ion sports (Ding et al., 2021). Different from participation in daily fitness and exer-
cise, consumer stickiness is stronger among fashion sports participation because
the ability to boost consumption is stronger, and the industry attributes are more
 Service Provider Selection 237

distinctive. When a consumer chooses to participate in a particular daily workout,

s/he is very likely to be restricted to stay within the confines of her/his city and
is very likely to remain faithful to a certain sport in a fixed facility. Conversely,
participation in fashion sports, such as events that relate to snow and ice, moun-
tain sports, water sports, automobiles and motorcycles, air sports, extreme sports,
equestrian, etc. usually relies on the diversity of landscape resources, such as
mountain land, water, and air area, wetland and wharfs, as well as forest land
resources. The development of fashion sports also relies on diverse landscape
resources. According to industry attributes and developing trends of various key
investment projects, a variety of aspects should be focused on. Indoor-leading
sport facilities should pay more attention to the selection of suppliers who can
improve internal service performance within the facilities, while outdoor-leaning
sport facilities need more supplementary service providers to cooperate with, such
as restaurants, overnight accommodations, transportation, entertainment, travel,
and retail in order to build a sport-core pooling of interest’s mechanism.

References
Beamon, B. M. (1999). Measuring supply chain performance. International Journal of
Operations & Production Management, 19(3), 275–292.
Billhardt, H., Hermoso, R., Ossowski, S., & Centeno, R. (2007, March). Trust-based ser-
vice provider selection in open environments. In Proceedings of the 2007 ACM sympo-
sium on Applied computing (pp. 1375–1380). ACM.
Blair, M. M. & Stout, L. A. (1999). A team production theory of corporate law. Virginia
Law Review, 85(2), 247–328
Chen, B., Chu, J., & Chen F. (2012). Research on government purchasing public employ-
ment service provider selection based on intuitionistic fuzzy multi-attribute decision.
Chinese Journal of Management Science, 20, 499–505.
Chen, Q., and Qi, F. (2009). An analysis of the index and the process of supplier selection—
The research based on the multinational companies’ purchase in China. Journal of
Shanxi University Philosophy &Social Science, 32(4), 42–48.
China’s State Council. (March 22, 2018). Guidelines on promoting the development of
region-wide Tourism. Retrieved from http://www.gov.cn/zhengce/content/2018-03/22/
content_5276447.htm
Cho, D. W., Lee, Y. H., Ahn, S. H., & Hwang, M. K. (2012). A framework for meas-
uring the performance of service supply chain management. Computers & Industrial
Engineering, 62(3), 801–818.
Churchill, G. A., Jr. (1979). A paradigm for developing better measures of marketing
constructs. Journal of Marketing Research, 64–73.
Cooper, M. C., Lambert, D. M., & Pagh, J. D. (1997). Supply chain management: more than
a new name for logistics. The International Journal of Logistics Management, 8(1), 1–14.
CPC central committee & Chinas State Council. (October 25, 2016). The Outline of
the Healthy China 2030 Plan. Retrieved from http://www.gov.cn/zhengce/2016-10/25/
content_5124174.htm
Dickson, G. W. (1966). An analysis of vendor selection systems and decisions. Journal of
Purchasing, 2(1), 5–17.
238 Y. Ding and Y. Y. Wang

Ding, Y., & Pan, S. (2018). The logical motivation and path of sports complex transforma-
tion and development—Based on the “people-centered” sport value orientation. Journal
of Shanghai University of Sport, 42(6), 32–35.
Ding, Y., Wang, Y. Y., & Baker, T. A. (2021). The application of residual control rights
theory to the entrusted operating agreements of multi-purpose sports complexes
in China—An exploration based on content analysis. International Journal of Sport
Management. 22(1), 27–54.
Ding, Y., & Zhang, L. (2019). Research on consumer service demand of multi-purpose
sports complex—Based on consumer survey of Jiangsu Province. Sports & Science, 2,
105–114.
Donaldson, T., & Preston, L. E. (1995). The stakeholder theory of the corporation:
Concepts, evidence, and implications. Academy of Management Review, 20(1), 65–91.
Dou, Y., Bai, S., & Chu, J. (2016). Coordination strategies of investing and financing
in supply chain based on suppliers’ repurchase incentives. Management Review, 28(6),
205–215.
Ellram, L. M., Tate, W. L., & Billington, C. (2004). Understanding and managing the
services supply chain. Journal of Supply Chain Management, 40(3), 17–32.
Ensign, P. C. (2001). Cross-border acquisitions in response to bilateral/regional trade lib-
eralization. Transnational Corporations, (1), 89.
Fisher, M. L. (1997). What is the right supply chain for your product? Harvard Business
Review, 75, 105–117.
Freeman, R. E. (1984). Strategic management: A stakeholder approach. Boston, MA: Pitman.
GASC. (July 13, 2016). The 13th Five-Year Plan of the Sport Industry. Retrieved from
http://www.sport.gov.cn/n316/n340/c733605/content.html
GASC. (May 11, 2017). Notice of the General Office of the State Administration of Sport
on promoting the construction of towns with sports and leisure features. Retrieved from
http://www.sport.gov.cn/n316/n336/c802334/content.html
GASC. (May 5, 2016). “The 13th Five-Year Plan” of Sport Development. Retrieved from
http://www.sport.gov.cn/n316/n340/c723004/content.html
GASC. (November 3, 2020). The selection results of typical cases of sports service com-
plex. Retrieved from http://www.sport.gov.cn/n316/n336/c968302/content.html
GASC. (October 10, 2020). Opinions of the General Office of the State Council on
strengthening the construction of fitness venues and facilities for the whole people and
developing mass sports. Retrieved from http://www.gov.cn/zhengce/content/2020-10/10/
content_5550053.htm
GASC. (October 11, 2020). During the 14th Five-Year Plan period, about 1,000 sports
parks will be built or renovated or expanded across the country. Retrieved from http://
www.gov.cn/zhengce/2020-10/11/content_5550334.htm
GASC. (September 2, 2019). Notice of the General Office of the State Council on print-
ing and distributing the outline for building a strong sports country. Retrieved from
http://www.gov.cn/zhengce/content/2019-09/02/content_5426485.htm
Guariguata, L., Whiting, D. R., Hambleton, I., Beagley, J., Linnenkamp, U., & Shaw, J.
E. (2014). Global estimates of diabetes prevalence for 2013 and projections for 2035.
Diabetes Research and Clinical Practice, 103(2), 137–149.
Handfield, R. B., & Nichols, E. L. (1999). Introduction to supply chain management. Upper
Saddle River, N.J.: Prentice Hall
Handfield, R., Walton, S. V., Sroufe, R., & Melnyk, S. A. (2002). Applying environmen-
tal criteria to supplier assessment: A study in the application of the analytical hierarchy
process. European Journal of Operational Research, 141(1), 70–87.
 Service Provider Selection 239

Hartley, J. L., Zirger, B. J., & Kamath, R. R. (1997). Managing the buyer-supplier interface
for on-time performance in product development. Journal of Operations Management,
15(1), 57–70.
He, D. (2003). An introduction of international purchasing in major global enterprises.
China Materials Distribution. 21, 39.
Hendry, J. (2001). Missing the target: Normative stakeholder theory and the corporate
governance debate. Business Ethics Quarterly, 159–176.
Hu, L., (2005). Application of the improved TOPSIS based on entropy for choosing the
business outsourcing suppliers. Logistics Management, 8, 60–62.
Huang, Z., Cai, W., & Zhang, B. (2013). Performance evaluation of logistics service sup-
ply chain from stakeholder theory perspective: Case of Chuanhua Logistics. Journal of
Xidian University, 5, 1–10.
Huang, Y., Yi, D., & Jie, M. (2016). Employee turnover of small and micro enterprises in
China: Present situations, impacts and determinants. Management World, 12, 77–89.
Jensen, M. C. (2010). Value maximization, stakeholder theory, and the corporate objec-
tive function. Journal of Applied Corporate Finance, 22(1), 32–42.
Kaliszewski, I. (2012). Quantitative Pareto analysis by cone separation technique. New York,
NY: Springer.
Kan, S. & Chen, R. (1998). Manufacturer supplier relationship in JIT environment: JIT
purchasing and supply in Chinese automobile industry. Journal of Industrial Engineering
and Engineering Management. 3, 46–51.
Kang, Y. (2007). A research of commercial real estate management model based on inte-
grated supply chain theory. Economic Survey. 3, 69–72.
Kannan, V. R., & Tan, K. C. (2002). Supplier selection and assessment: their impact on
business performance. Journal of Supply Chain Management, 38(4), 11–21.
Ko, Y. J., & Pastore, D. L. (2005). A hierarchical model of service quality for the recrea-
tional sport industry. Sport Marketing Quarterly, 14(2). 84–97.
Lee, S., & Ross, S. D. (2012). Sport sponsorship decision making in a global market: An
approach of Analytic Hierarchy Process (AHP). Sport, Business and Management: An
International Journal, 2(2), 156–168.
Li, J. (2006). Research on supplier selection based on supply chain partnership. Journal of
Beijing Technology and Business University, 21(6), 40–45.
Li, J., & Wang, Y. (2015). Evaluation of third party reverse logistics suppliers based on
Fuzzy-QFD. Logistics Technology.10(2), 165–168.
Liang, T., Liu, H., & Li, N. (2011). An empirical analysis of the matching degree between
industrial structure and talent Structure – A case study of Liaoning Province. Journal of
Industrial Technological Economics, 12, 80–85.
Liu, X., Hou, G., & Zhang, D. (2003). Analysis of cooperation matching degree in inter-
national joint ventures. Nankai Business Review, 3, 61–63.
Liu, P., & Zhang, X. (2011). Research on the supplier selection of a supply chain based on
entropy weight and improved ELECTRE-III method. International Journal of Production
Research, 49(3), 637–646.
Mentzer, J. T., DeWitt, W., Keebler, J. S., Min, S., Nix, N. W., Smith, C. D., & Zacharia, Z.
G. (2001). Defining supply chain management. Journal of Business logistics, 22(2), 1–25.
Oliver, R. K., & Webber, M. D. (1982). Supply-chain management: logistics catches up
with strategy. Outlook, 5(1), 42–47.
Roy, S., Sivakumar, K., & Wilkinson, I. F. (2004). Innovation generation in supply chain
relationships: a conceptual model and research propositions. Journal of the Academy of
Marketing Science, 32(1), 61–79.
240 Y. Ding and Y. Y. Wang

Saaty, T. (1980). The analytic hierarchy process. New York, NY: McGraw-Hill.
Saaty, T. (1982). Decision-making for leaders. Belmont, CA: Wadsworth.
Saaty, T. L. (2008). Decision making with the analytic hierarchy process. International
Journal of Services Sciences, 1(1), 83–98.
Saaty, T. L., & Ozdemir, M. S. (2003). Why the magic number seven plus or minus two.
Mathematical and Computer Modelling, 38(3-4), 233–244.
Saccani, N., Johansson, P., & Perona, M. (2007). Configuring the after-sales service supply
chain: A multiple case study. International Journal of Production Economics, 110(1-2),
52–69.
Sengupta, K., Heiser, D. R., & Cook, L. S. (2006). Manufacturing and service supply chain
performance: a comparative analysis. Journal of Supply Chain Management, 42(4), 4–15.
Simchi-Levi, D., Kaminsky, P., & Simchi-Levi, E. (2008). Designing and managing the supply
chain: concepts, strategies, and case studies (3rd ed.). Boston, MA: McGraw-Hill/Irwin.
Song, D. & Huang, W. (2010). Evaluation of producer service supplier in the perspec-
tive of service supply chain. Journal of Wuhan University of Technology (Information &
Management Engineering), 3, 473–477.
Stentoft Arlbjørn, J., Vagn Freytag, P., & de Haas, H. (2011). Service supply chain man-
agement: A survey of lean application in the municipal sector. International Journal of
Physical Distribution & Logistics Management, 41(3), 277–295.
Teknomo, K. (2006). Analytic hierarchy process (AHP) tutorial. Revoledu.com, 1–20.
Wagner, S. M., & Hoegl, M. (2006). Involving suppliers in product development: Insights
from R&D directors and project managers. Industrial Marketing Management, 35(8),
936–943.
Wang, C., & Wang, G. S. (2005). Analysis of strategic compatibility in corporation
restructuring. Soft Science, 25(2), 90–93.
Wang, Q. T., Zhao, Y., Guo, H. T., & Zheng, Z. G. (2013). Analysis of Large-scale Sports
Event Logistics and Supply Chain Management. Logistics Technology, 11, 200–202+209.
Wang, Y., Wallace, S. W., Shen, B., & Choi, T. M. (2015). Service supply chain manage-
ment: A review of operational models. European Journal of Operational Research, 247(3),
685–698.
Wasti, S. N., & Liker, J. K. (1997). Risky business or competitive power? Supplier involve-
ment in Japanese product design. Journal of Product Innovation Management, 14(5),
337–355.
Wynstra, F., & Ten Pierick, E. (2000). Managing supplier involvement in new product
development: a portfolio approach. European Journal of Purchasing & Supply Management,
6(1), 49–57.
Xiao, Z. & Liao, L. (2010). Non-financial stakeholders and company capital structure
selection: A dynamic adjustment model. China Industrial Economics, 10, 175–189.
Xu, D. Y., & Han, B. T. (2015). The model and empirical study on the match between
policy supply and demand: Based on innovation and entrepreneurship policy in Beijing.
Studies in Science of Science, 12(4), 1787–1796.
Xue, H., & Mason, D. S. (2019). Stadium games in entrepreneurial cities in China: A
state project. Journal of Global Sport Management, 4(2), 185–209.
Yan, X., Sun, L. & Wang, K. (2005). Research on performance evaluation and characteris-
tics in logistics service supply chain. China Mechanical Engineering, 6, 969–974.
Yang, J. (2006). Mechanism of inter-firm trust in supply chain and its influence on coop-
eration: Based on the research of manufacturing enterprises. Unpublished doctoral dis-
sertation, Zhejiang University, Hangzhou, China.
 Service Provider Selection 241

Yang, Z., Xu, Q., & Peng, W. (2008). A Study on the method for supplier selection based
on principal component analysis and TOPSIS. Journal of Information, 11, 7–10.
Ye, H. (2008). Research on investment and financing of water pollution treatment.
Statistics & Decision, 2, 57–59.
Yu, H. (2011). Study on operational optimization of rural circulation supply chain based
on contract coordination mechanisms of service supply chains. Unpublished doctoral
dissertation, Zhejiang University, Hangzhou, China.
Zeng, L., Yan, Z., & Zhu, Q. (2007). The application of AHP in the evaluation and selec-
tion of suppliers for transportation. Journal of Guangdong Communications Polytechnic.
6(1), 63–65.
Zhang, J., & Zhao, Q. (2018). Transformation and upgrading of the business model of the
circulation supply chain driven by ‘new retail’. Journal of Business Economics, 11, 5–15.
Zhen, Q., Yu, D. Sun C., & Zhang, G. (2017). Research on allocation of urban basic
education facilities based on supply and demand matching: A case study of Hefei City.
Journal of East China Normal University, 1, 133–138.
Zhu, Y. (2004). Empirical studies on the integration of the agricultural supply chain and
logistics. Unpublished doctoral dissertation, Nanjing Agricultural University, Nanjing,
China.
Chapter 13

COVID-19 and Economic

Impacts of Major Sport Events
Lidija Petrović, Miaad Ghasemzadeh, and
Christopher Hautbois

Introduction
The understanding of economic concepts applicable in the sports industry has
advanced and is now much broader in scope. The distinctive characteristics of the
sports industry suggest that the “sport as business” comprises numerous aspects
aimed at general development and regional specialism addressing, among other
issues, aspects of managing sports events (Beech & Chadwick, 2004; Desbordes
& Falgoux, 2007; Desbordes, 2012; Desbordes & Richelieu, 2012; Dolles &
Söderman, 2011; Humphreys & Maxcy, 2007; Petrović & Desbordes, 2012,
Petrović et al., 2015; Trenberth & Hassan, 2011) and their beneficial impacts –
some crucial determinants of which are economic (e.g., Andreff & Szymanski,
2006; Rodríguez et al., 2006; Sandy et al., 2004). Management of contemporary
sports events involves an application of methods and models that are also used in
commercial business as well as in public and not-for-profit organizations (Petrović,
2012). In this regard, over the past several decades, much academic effort has been
invested in the development of methods and models which measure the economic
impacts of different major sports events (Andersson et al., 2008; Blake, 2005; Burns
et al., 1986; Baade & Matheson, 2002; Desbordes, 2007; Dwyer et al., 2000; Getz,
1994, 1997; Gratton et al., 2000, 2006; Kurscheidt, 2000; Matheson, 2002, 2006;
Oldenboom, 2006; Preuss 1999, 2005, 2010; Syme et al., 1989; Ritchie, 2000;
Solberg et al., 2002; Solberg & Preuss, 2007).
Discussions on the methodology and their research scope that may be found
in the sports management literature are worth special mention (Crompton, 1995,
2006; Howard & Crompton, 1995; Porter, 1999). Research attempts have also
been made to apply the Balanced Scorecard approach (Kaplan & Norton, 1992;
1996a; 1996b) to sports events evaluation (Gratton et al., 2006) since the bene-
fits associated with events are far-reaching and not confined solely to the anal-
yses of economic impacts widely used thus far. The scorecard developed aimed
at organizing the effects of a major sports event across four Balanced Scorecard
perspectives, pulled together important elements of the hosts’ agenda, but it did
not claim to cover all the possible impacts (Dolles & Söderman, 2008). The var-
ious assessments are much more than just simple evaluation tools, although they

DOI: 10.4324/9781003302476-13
 COVID-19 and Economic Impacts 243

must not lose their impartiality. For this reason, some of the assessments should
be conducted by external institutions or consultants whose independence and
lack of direct involvement could be a guarantee of quality of the results presented.
The evaluation can be considered from different angles because there are as many
ways of evaluating the sports event as there are points of view and stakeholders
(Desbordes, 2012).
This is a hard time for major sports events and the sports economy. Generally
speaking, hosting a major international sports event is a costly affair for the
organizing country, as Matheson et al. (2012) mentioned. Further, the unprec-
edented COVID-19 pandemic and the lockdowns associated with attempts to
stop its spread have made a great impact on both elite and grassroots sport. Elite
sport teams and organizations have suffered immediate financial losses due to the
absence of live spectators. As an example, The Rugby Football Union (RFU) has
suggested that with no spectators attending the recent Nations Cup or the Six
Nations, they would lose close to £60 million (BBC, 22 September 2020). It is not
only elite sport that has been impacted, but grassroots sport, leisure clubs, and
organizations, including gyms, swimming pools, golf courses, and other activities,
have also been hit financially (The Guardian, October 13, 2020). Equally, gate and
ticket receipts constitute a large part of the revenue football teams receive and
the prolonged pandemic has led to many in the English football pyramid, one
the most lucrative in the world, struggling financially (The Week, September 22,
2020). Therefore, planning, organizing, and staging a major sports event is not
a free lunch; as such, an event requires different investments in infrastructure,
facilities, security, and other services (Petrović, 2016).
The COVID-19 pandemic, as the greatest global health crisis of the last decades,
has spread to almost all countries of the world and has affected all aspects of peo-
ple’s lives, including sport events. The pandemic, which is characterized as one of
the more significant human experiences of the twenty-first century (Galvani et al.,
2020), has caused the most significant disruption to the worldwide sports calendar
since World War II. The Olympics, for the first time in the history of the modern
games, have been postponed until 2021, World Championships and World Cups for
virtually every sport have been canceled or postponed, and the calendars for most
major sports have been severely disrupted with no clear indication yet of how or
when this will be resolved. Among the sports events that have been postponed are
the running events, those at the international, regional, or local level. At the same
time, serious concerns arose regarding the resumption of these events, the rules of
their conduct but also the intention of the runners to participate in them (Burrows
& Flynn, 2020). The sports events’ organizers, like every other business in tourism
and cognate industries, will have to provide evidence of the implementation of
satisfactory health and safety measures to convince customers to come back again
(Seraphin, 2020) and safeguard the health of athletes, participants, visitors, and
others involved (Burrows & Flynn, 2020).
Different aspects affect the planning and staging a major international sports
event; with regards to the sports event industry practice and for the purposes
244 L. Petrović, M. Ghasemzadeh, and C. Hautbois

of this paper, the term “major event” is used to follow the definition as sug-
gested by Jago and Shaw (1998). The chapter reviews the three main types of
economic analyses addressing measuring the economic impacts of major sports
events, namely, input–output analysis in the second section, cost-benefit analysis
in the third section, and computable-general-equilibrium modeling in the fourth
section – all three having its own and specific characteristics. This classification
of approaches to measuring the economic impact is followed by further explana-
tion and analysis of the Balanced Scorecard framework as an evaluation model
in the fifth section, modified and adapted for the needs of a major sports event
evaluation. Finally, concluding remarks and some future directions are provided
in the sixth section.

Input-Output Analysis
According to Andersson et al. (2008), “one of the most popular models, the
input-output (I-O) analysis describes financial flows; more specifically money that
flows into an economy is a positive flow, and money that flows out is a negative
flow” (p. 165), better known as leakage. Used commonly by researchers and ana-
lysts to estimate economic impacts, I-O analysis measures inter-industry linkages
in the national economy to explain the transactions between businesses and final
consumers. I-O analysis uses comprehensive matrices reflecting inter-industry
transactions. Duncombe and Wong (1998) state, “the heart of I-O analysis is the
transaction table” (p. 168); indeed, the original I-O tables were constructed by
W.W. Leontief in the 1930s, and his research and development in this field earned
him the Nobel Prize. I-O tables estimate three sets of effects – direct, indirect,
and induced impacts resulting from the final demand institutions (e.g., households,
governments, inventory/capital purchases, exports, and inter-institutional trans-
fers). For each of these sets, multipliers can be calculated that divide the gross
value added or “net output” (GVA), gross domestic product (GDP), or import impact
by the level of spending that drives that impact.
The multiplier is an estimate of how much additional economic activity will
result from an investment in the economy; it is called the multiplier due to that
total impacts are larger than the initial, direct impacts. It is the value which can
lead to inaccurate, exaggerated, and spurious inferences (Crompton, 2006) and
the widespread misuse of the concept in terms of “turnover multiplier” (Andersson
et al., 2008), inflates direct turnover effects to 10–90 percent higher than the
original direct effects. The value of the multiplier is the most sensitive param-
eter in an economic impact analysis. Sandy et al. (2004) call it the “multiplier
game”. Numbers found in consultants’ studies on regional multipliers vary from
1.2 to 5 (Matheson, 2002), with employment multipliers being somewhat lower.
Furthermore, all values above 2 are highly suspect and should be considered with
care. I-O tables assume constant returns to scale, and balance supply and demand
in the production sector. They do not account for supply side variables such as
price, technology, and industry competition, assuming full output homogeneity
 COVID-19 and Economic Impacts 245

within an industry sector. A complete set of account for an economy, typically for
a certain year, is shown in I-O tables.
This technical note describes in details the methodology developed for
assessing the daily economic costs of sports events for mitigating the effects
of COVID-19. Economic impact of sports events can be defined as the net
change in an economy resulting from a sports event. The change is caused by
activity involving the acquisition, operation, development, and use of sport
facilities and services (Lieber & Alton, 1983). These in turn generate visi-
tors’ spending, public spending, employment opportunities, and tax revenue.
Specifically, the economic impacts of expenditure are composed of direct,
indirect, and induced effects. Direct effects are the purchases needed to meet
the increased demand of visitors for goods and services. Indirect effects are the
ripple effect of additional rounds of re-circulating the initial spectators’ spend-
ing. Induced effects are the increase in employment and household income
that result from the economic activity fueled by the direct and indirect effects
(Howard & Crompton, 1995).
The basic requirement of an economic impact analysis is to measure the total
change of the city structure due to staging a sports event. Therefore, determining
resources entering, leaving, rejection or exchange is crucial, as it is determining
new or lost infrastructural projects due to investments being diverted (Preuss,
2010). In order to conduct an economic impact analysis of a major sports event,
the following stages must be undertaken. Firstly, the effect of the event on spend-
ing by organizations such as the Local Organising Committee (LOC) as well
as by individuals must be calculated. Spending by the LOC includes infrastruc-
tural spending in the pre-event period as well as spending during the period of
the event itself. Spending by individuals includes the transport, accommodation,
food, and entertainment expenditures of spectators as well as athletes, officials,
and media representatives; although some of these categories have accommoda-
tion, food, and/or transport provided through the LOC’s budget. Care must be
taken to differentiate spending by residents from spending from non-residents and
also to take account of spending that is diverted away from the host city. Secondly,
this expenditure must be categorized by product (goods and services), and thirdly,
a model must be used to calculate how this spending translates into income and
employment.
It should be noted that GVA is sometimes also termed “net output”, which
is very important regarding the true impact of an event, as results for “output”
meaning gross output are often given in the event’s literature, which are then mis-
interpreted as being results for net output or GVA. Gross output is a measure of
revenue and is in most industries significantly larger than GVA; therefore, impact
calculations of gross output changes are usually much higher than earnings-based
estimates through GVA. It is also important to underline that gross output and
measures of impact based on it can have no interpretation as “benefit”. GVA and
GDP based measures are a suitable way of approximating the benefit of events as
explained by Blake (2005).
246 L. Petrović, M. Ghasemzadeh, and C. Hautbois

Reasons to Conduct the I-O Analysis of Major Sports Events

A number of economic impact analyses have been conducted of major sports
events (including the largest ones, such as the Olympic Games, FIFA World
Cups, UEFA European Championships) using I-O methods. Hosting a sports
event has revealed a number of benefits in our communities, of which the rea-
sons like increasing community visibility and enhancing community image are
all common and acceptable postulations. However, there is a doubt that sports
events that utilize public subsidies always bring positive economic benefits into
communities. Therefore, the main reasons for an I-O analysis of sports events
may be seen as follows: First, due to that many sports events in communities
are usually financed by the public tax support, I-O analyses continue to be an
important public relations tool for local government. Second, there is a doubt that
sports events may actually help developing a community in relation to its econ-
omy. Therefore, accurate estimations may be proposed and the results reported
to community members. Third, as a sport is not just an entertainment but the
industry, the results of the I-O analyses may be a cornerstone to develop many
related businesses in communities. Finally, positive or negative economic results
of sports events may be an important method to determine communities’ draft
budget for the coming year.

Problems of the I-O Analysis of Major Sports Events

This method of evaluation of major sports events is very controversial due to
its subjective aspects. There are several problems mentioned in the literature on
using I-O analyses in the context of sports event evaluation so far. The first prob-
lem may be seen in difficulties in determining the “right” data and framework,
the lack of established methodology, and the mistakes accompanying such anal-
ysis that explain various results of an economic impact analysis (Preuss, 2010).
Establishing the method of data collection – bottom-up or top-down – as well
as the context and time frame for analysis are only a few of the issues requiring
attention. Second, not only is a static approach (comparing an event case to a
without event case scenario) possible, but a dynamic approach comparing the
event case to a control case or an econometric approach comparing the event
case to a reference case is possible as well. With smaller sports events and less
direct impacts, it is increasingly difficult to find the right data for collection. For
this reason, the top-down approach is limited for impact measurement, especially
when considering larger host regions. In contrast to this, the bottom-up approach
remains the preferable option for economic impact assessment; however, it is not
without its own challenges. The third problem is in use of different and conflicting
concepts of the multiplier itself (Howard & Crompton, 1995).
A danger in the multiplier and the way it is presented in research reports
intended for policy makers is that its basic concept and application are decep-
tively sample. This means that economic impact studies are primarily used by
 COVID-19 and Economic Impacts 247

consultants hired by sport entrepreneurs and boosters to demonstrate the value

of a proposed sports event (Jonson & Sack, 1996). Forth problem is the inclusion
of local spectators, time-switchers, and casuals in the study. An economic impact
analysis attributable to a sports event should include only new cash flow injected
into an economy by visitors and other external businesses such as the media,
banks, and investors from outside the community. In addition, due to the fact
that expenditures of time-switchers and casuals would occur without the event,
impacts of their expenditures should be excluded from the economic impact anal-
ysis. The fifth problem is that an economic impact analysis conducted by hired
consultants from a political milieu usually estimates positive impacts only, which
means benefits both economically and socially. It never measures substantial eco-
nomic costs and/or potential social problems. Considering the side of economic
impact, only gross benefits rather than net benefits are measured and reported.
In the case of non-economic impacts, negative social impacts, including traffic
congestion, vandalism, environmental pollution, disruption of residents’ lifestyle,
etc., are rarely reported. And finally, sixth problem, economic impact analyses are
very subjective, depending on researchers to trust their results. Although some
models and mathematical formulas for conducting economic impact analyses
have been developed and used, results and their interpretation could be changed
based on the intent of researchers and the unrealistic expectations of proponents.

Cost-Benefit Analysis
The COVID-19 pandemic leads governments around the world to making sport
decisions about the cost of saving lives and the restrictions of that. Applying
cost-benefit analysis is one of the methods to help the sports industry and econ-
omy. The cost-benefit analysis (CBA) is an alternative method to economic
impact analyses. It has a different approach: it helps to decide whether hosting
an event will improve the welfare of the community in the host region by com-
paring all relevant costs and benefits in monetary terms. Cost-benefit method is
considered the proper way to estimate economic consequences, as Jeanrenaud
(2006) explains.
The CBA is explicitly designed for determination of the total “value” of
the sports event relative to a consistent set of goals, both initial event-related
expenditures and their induced impacts that have to be categorized according to
their effects on those goals. Such an evaluation method provides a clear struc-
ture that helps in classifying the repercussions of the project under examination.
The effects themselves are to be distinguished in direct and indirect and tangible
and intangible costs and benefits. That is, CBA comprehends not only quanti-
tative but also qualitative information which is impossible or difficult to express
in a money equivalent. The qualitative part of the CBA provides an additional
perspective on the problem of evaluation, although consistently embedded in a
common methodological framework. Nevertheless, the core of the CBA is the net
present value (NPV), as a highly aggregated single figure that depicts all negative
248 L. Petrović, M. Ghasemzadeh, and C. Hautbois

and positive quantifiable impacts of the project (i.e., sports event). NPV is com-
puted by summing up all discounted net benefits, which occur over time within
a defined planning period. According to the Kaldor–Hicks welfare criterion, a
potential part to optimality, the NPV should at least exceed zero to reach an
overall positive evaluation (e.g., Preuss, 1999).
The main characteristics of the CBA may be seen as follows: (i) It is a decision-
support method which address choice problems in order to raise efficiency of a
project or actions; (ii) It is guided by the evaluation of effects or actions relative
to a hierarchical set of goals and objectives; (iii) It is value-oriented, as CBA
assesses the NPV of an economic contribution to GDP over time; (iv) It is an
approach not precisely defined – it rather provides a certain structure or model,
which allows identifying and considering key factors and variables relative to
which effects and actions, respectively, are to be evaluated and/or improved
by applying the concept; (v) and Ultimately, both CBA aims at applying basic
micro-economic principles to a well-defined object of examination in order to
support rational decision-making on complex choice problems (Kurscheidt,
2000).
Considering the major sports events as public projects, CBA plays the role of
a forecasting and a feedback instrument that allows an appropriate evaluation
in the process of strategic management. The result of a CBA is the qualitative
and quantitative assessment of the efficient action relative to alternatives and
thereby establishing or raising “allocative efficiency”. The outcome of the strate-
gic management is emerging from “hierarchy of strategic intent” (Miller, 1998) to
enhance the so-called production efficiency of the respective organization. In par-
ticular, CBA is suitable for examining the repercussions of specific action plans
on the overall result. It helps to identify and then to prevent possible trade-off
effects that might be induced since all the measures and variables work in the
same direction. Such counter-effects can be visible by CBA as it always shows
the total picture and critical values of the project’s impacts. Last but not least,
the structure of the CBA provides a deep understanding of the driving forces and
interrelationships of project effects. Thus, CBA itself can be a tool for indicating
efficient strategies that direct resources to a productive use.
The outcome of a CBA may be found in the main types of benefits and costs
that may be identified (Dwyer & Forsyth, 2009). On the one hand, the benefits
comprise payments to event organizers, resident consumer benefits, surpluses
to state business, and surpluses to labor; on the other hand, there are three
main sets of costs to be differentiated, i.e., major categories of organizer’s costs
(event management and staging, recurrent engineering, marketing/promotion,
and catering, and administration), other event-related costs incurred by gov-
ernment agencies (payments in respect of roads and traffic authority, police,
fire brigade, etc.), as well as social and environmental costs (noise, congestion,
disruption to resident lifestyle, impact on destination image, etc.). Therefore,
CBA should be used to account for the wider social and environmental impacts
of sports events.
 COVID-19 and Economic Impacts 249

Computable-General-Equilibrium Modeling
The COVID-19 pandemic has caused a strong discontinuance in the economy
worldwide. In many developing and developed countries, the sports economic
sector is a major source of occupation and revenue. Without this crucial section,
most of the countries might experience a dramatic decline in event revenue and
increase in rate of club bankruptcy. The computable general equilibrium model
may help to assess the implications of the COVID-19 crisis on the sports eco-
nomic sector. Computable-general-equilibrium (CGE) modeling has now largely
supplanted more traditional forms of economic analysis, such as CBA, and less
reputable methods, such as I-O analyses. They consist of several sub-models,
which describe various markets and economies, and a fundamental principle of
this evaluation technique is that markets and economies are treated as interde-
pendent (Andersson et al., 2008).
Generally speaking, CGE models are used in a wide variety of economic areas
such as international trade, free trade areas and customs unions, agricultural pol-
icy, economic development, and environmental policy. Recently, they have also
begun to be used in the field of the economic impact analysis of sports events,
where they are replacing I-O analyses, which are now seen as the “old” method.
As an example, the main analyses of the Sydney 2000 Olympic Games as well as
the 2010 FIFA World Cup South Africa have been conducted with CGE mode-
ling techniques. The main difference between I-O and CGE methods is that key
relationships that I-O analyses ignore are included in CGE models. The incorpo-
ration of these changes means that a CGE method is more complex than an I-O
analysis but also that it measures impacts more accurately. That is, I-O models
can measure all positive impacts of an event but are incapable of modeling most
of the negative impacts, so they consistently overestimate the impact of events.
Therefore, CGE modeling gives more realistic results.
However, most economists agree upon that I-O models suffer from several
major flaws, including, in the first instance that they impose no constraint on
the amount of extra income that can be earned by labor or capital. In contrast
to this, CGE models impose constraints on income and expenditure for private
households and separately for the government (the value of income must equal
expenditure) with additional constraints of a much higher level of complexity to
necessary model prices and wages and the quantity variables responding to them.
Two main differences emerge in the way that CGE models and I-O analyses pre-
dict the effects of a sports event, namely:

1 The effects of changes in demand have different effects in these two types
of models. I-O analyses capture the initial effects of sports spending plus the
indirect, and if included, the induced effects. The “multiplier” of sports spend-
ing on GDP is therefore fairly high, moreover, if induced effects are included,
can be greater than one, including (i) the effects of higher prices “crowding-
out” demand, and more significantly (ii) the movement of resources into
250 L. Petrović, M. Ghasemzadeh, and C. Hautbois

sport-related industries from other industries, with consequent falls in the

output of other industries, particularly in other exporting industries, which
have much lower multiplier effects (e.g., Adams & Parmenter, 1995; Blake
et al., 2001).
2 The second way in which CGE models differ in the impacts that they will
predict for sports events is that construction expenditures are not necessar-
ily positive. A CGE modeling requires income-expenditure conditions to
be met; therefore, the spending on construction must be paid for. The time
dimension should also be included in the effects of the constructed infra-
structure being available after completion (e.g., government that paid for the
construction might rent out the built infrastructure or receive income from
its sale. If the value of constructed capital exactly pays for its construction
costs, there may be an initial net zero effect on GDP).

Balanced Scorecard Approach

The new pandemic disease has a direct influence on sport events and organi-
zations. The assessment of sports organizations in a pandemic situation has a
different approach and must have certain factors. Many organizations don’t have
the means or resources to quickly and effectively assess the impact of the pan-
demic on their organization. However, many tools and methodologies exist for
the organizations that can be used to help them assess the impact with great
success. One of the implementations is well-known as the Balanced Scorecard
(BSC), which is a framework that organizations can use to verify that they have
established both strategic and financial controls to assess their performance. This
technique is most appropriate to be used when dealing with business-level strate-
gies, although it can also be successfully used with the other strategies organiza-
tions may choose to implement at different levels – corporate, unit, cooperative,
and international. Originally proposed by Kaplan and Norton, the BSC can help
to get a good perspective for the complete evaluation of a company’s activities.
While the initial focus and application of the BSC has been in the for-profit–
private sector, as emphasized by Kaplan and Norton (1996a), the opportunity for
the scorecard to improve the management of public and not-for-profit organiza-
tions is, if anything, even greater.
Since traditional metrics were not offering a full understanding being in most
cases outcome focused instead of being process orientated, both academics and
practitioners were urged to develop better metrics that could ease the process
of strategic planning and control. The major inadequacies of traditional met-
rics that may be found in the literature are as follows: (i) traditional accounting
measures of performance are inadequate for strategic decisions; they do not link
non-financial metrics to financial numbers (e.g., Kaplan & Norton, 1992); (ii)
they are backward-looking, providing little information on root causes; (iii) they
report functional not cross-functional processes; (iv) they rarely consider intangi-
ble assets; and (v) they rarely estimate the value created.
 COVID-19 and Economic Impacts 251

Poorly designed performance measurement systems (mostly based on account-

ing measures) and the pressure for short-term financial performance also caused
companies to reduce spending on new product development, process improve-
ments, human resource development, information technology, and customer
and market development. It was to avoid such “distorted” incentives when the
deployment of the BSC was suggested. As a performance management tool, the
BSC has been designed to assist management in aligning, communicating, and
tracking progress against ongoing business strategies, objectives, and targets. The
BSC is unique in that it combines traditional financial ratios with non-financial
measures to measure the health of the company from the four equally important
perspectives:

1 Financial: Measures the economic impact of actions on growth, probabil-

ity, and risk from shareholder’s perspective (net income, ROI, ROA, and cash
flow);
2 Customer: Measures the ability of an organization to provide quality goods
and services that meet customer expectations (customer retention, profitabil-
ity, satisfaction, and loyalty);
3 Internal Business Processes: Measures the internal business processes that
create customer and shareholder satisfaction (project management, TQM, and
Six Sigma);
4 Learning and Growth: Measures the organizational environment that fosters
change, innovation, information sharing, and growth (staff morale, training,
and knowledge sharing) (Bauer, 2004).

Although the focus of each perspective is distinctly different, there is a com-

mon thread of causality that provides a universal linkage among the four per-
spectives. The cause-and-effect chain is central to the BSC. It is the chain that
distinguishes the model from other approaches. Kaplan and Norton (1996a)
assume the following causal relationship: measures of organizational learning
and growth → measures of internal business processes → measure of the customer
perspective → financial measures. Therefore, the measures of organizational
learning and growth are the drivers of the measures of the internal business
processes. The measures of these processes are in turn the drivers of the meas-
ures of the customer perspective, while these measures are the drivers of the
financial measures. A good BSC should have a mix of outcome measures (= lag
indicators) and performance drivers (= lead indicators). (An example of a lag
indicator is increased turnover, while order execution time is a lead indicator.) Each
strategic area should have both lead and lag indicators, yielding two-directional
cause-and-effect chains; lead and lag indicators apply horizontally within the
areas and vertically between the areas. The causal paths from the measure indi-
cators on the scorecard should be linked to financial objectives. This procedure
implies that strategy is translated into a set of hypotheses about cause and effect
(Kaplan & Norton, 1996a, 1996b).
252 L. Petrović, M. Ghasemzadeh, and C. Hautbois

One of the first applications of the BSC to a not-for-profit social enterprise

occurred for the Massachusetts Special Olympics. The framework of the Special
Olympics BSC was virtually identical to that which was used for companies and
business units. In 1991, in the UK, the Audit Commission set out a series of indica-
tors to evaluate civic management. As a result of these efforts, the publication of a
report in 2000 (Audit Commission, 2000) stressed the usefulness of focusing on a
series of key objectives. Therefore, one of the ways of developing performance indi-
cators is to use the three “E” frameworks, i.e., dimensions of economy, efficiency, and
effectiveness; these three economic dimensions deal with a set of elements necessary
to obtain them. These elements are infrastructure investments and cost estimations
(cost), the human and material resources employed (resource), the services-processes
and single activities obtained (service), the results of the activities (output), and the
value that is desired-expected to be achieved (outcome). The aforementioned eco-
nomic concepts – economy, efficiency, and effectiveness – link two of the elements
(Audit Commission, 2000; Petrović, 2012).
Applied in this context, the BSC forms the platform for the evaluation frame-
work, assessing both direct and indirect effects of the performance determinants-
perspectives on the sports event, using the line management indicator framework,
the three “E” framework, as a basic evaluation tool. A performance measurement
system as an evaluation framework usually has a wide range of users who may
use the information in different ways. That is, in accordance with the level of
the information users (management & staff/local-national–public & stakehold-
ers/international-public & governing bodies), the cause-and-effect relationships
may be considered and an adapted version of Kaplan and Norton’s BSC may
be proposed with the modifications in the BSC perspectives, following Bauer’s
key indicator matrix initially developed for-profit companies and adapted for the
needs and purposes of a sports event evaluation. As the BSC approach includes
different areas of the organization, as well as showing the cause-and-effect rela-
tionships which can help to determine whether the various processes (in line
with the vision and strategy at the time of planning) achieve their objectives, the
basic structure of the modified BSC to meet the needs of the stakeholders of a
major sports event, is seen as shown in Figure 13.1.

Financial/Economic Impact Perspective

The modifications to be found in financial perspective are focused on the use
of the financial resources and designated funds employed in the planning and
organization of the sports event. Shareholders in the corporate context are
replaced here with stakeholders in the staging and implementation of the sports
event. Typical criteria of the profit companies in terms of an increase in financial
performance are considered redundant in this case; therefore, this perspective has
to show how resources are employed using cost estimations and revenue generated
and how are the same reflected on the sports event’s equity as well as the market-
ing of the sport event, important for the prospective organizer.
 COVID-19 and Economic Impacts 253

Figure 13.1 The Stakeholders’ Approach to Balanced Scorecard Sports Event’s Evaluation.

Customer/Socio-Cultural Impact Perspective

In this perspective, the BSC model has to consider those who benefit from the
staging of a sports event; therefore, it is reasonable to think more in terms of con-
sumers as participants and non-attendees instead of corporate customers. This
perspective has to show how cognitive indicators estimate attitude behavior and
satisfaction of participants, considering that the professional sport creates benefits
for the population in general.

Internal Business Processes/Service Impact Perspective

Internal business process perspective is replaced by the impact created from ser-
vices aimed at identifying both the key processes and single activities generated
during the sports event’s life-cycle. The processes should be monitored in order to
achieve desired results related to the event’s program (the most important indica-
tors are related to the program), possible disorganizations, strategy for the tickets’
sales, etc. A better co-ordination of the activities of the program could be accom-
panied by the co-operation with other institutions in their organization.

Learning and Growth/Environmental and

Territorial Impact Perspective
This perspective is similar to that found in the BSC of the profit company. The key
aspects are focused on continuing improvement of environmental and ecological
254 L. Petrović, M. Ghasemzadeh, and C. Hautbois

Table 13.1 M
 atrix of Key Indicators Considering the Perspectives of the Balanced
Scorecard for Major Sports Event Evaluation

Performance Measurements

Economic impact perspective Socio-cultural impact perspective

(STAKEHOLDER) (PARTICIPANT AND NON-ATTENTEES)
• Performance of financial resources • Product quality and innovation
(designated funds) • Consumer loyalty and satisfaction
• Cost estimations and savings in • Service quality, etc.
expenses
• Resource utilization
• Staff productivity, etc.
Service impact perspective Environmental and territorial impact
(KEY PROCESS AND SINGLE perspective
ACTIVITIES) (CONTINUING IMPROVEMENTS)
• Quality of processes and activities • Innovation and improvements’
related to the event’s program implementation, updating
• Productivity • Technological advances
• Efficiency of processes and services • Knowledge and information exchange, etc.
• Event cycle time, etc.

Source: Modified from Bauer (2004).

aspects via checking information on garbage generated from visitors and par-
ticipants, electricity consumption, physical damage of infrastructure, quality of
the air, level of noise, etc. In order to work successfully, the BSC has to be well
designed. That means it should integrate the key indicators into matrix, based on
the BSC, which reflects the desired “activity” to be evaluated. The matrix of key
indicators in Table 13.1, below considering the perspectives of the BSC model
for a major sports event evaluation, has been developed following Bauer’s (2004)
matrix initially intended for companies.

References
Adams, P., & Parmenter, B. (1995). An applied general equilibrium analysis of tourism in
a quite small, quite open economy. Applied Economics, 27(10), 985–994.
Andersson, T. D., Armbrecht, J., & Lundberg, E. (2008). Impact of mega-events on the
economy. Asian Business & Management, 8, 163–179.
Andreff, W., & Szymanski, S. (2006). Introduction: Sport and economics. In W. Andreff
and S. Szymanski (Eds.) Handbook on the economics of sport. Cheltenham, UK: Edward
Elgar.
Audit Commission (2000). On target-the practice of performance indicators. London, UK:
The Audit Commission.
Baade, R. A., & Matheson, V. A. (2002). Bidding for the Olympics: Fool’s gold? In C. P.
Baros, M. Ibrahimo & S. Szymanski (Eds.), Transatlantic sport: The comparative econom-
ics of North American and European sports (pp. 127–151). London: Edward Elgar.
 COVID-19 and Economic Impacts 255

Bauer, K. (2004). Key performance indicators: The multiple dimensions, DM Review,

October Issue.
BBC. (2020, September 22). London, England: Coronavirus: fans may not be able to return
to sporting events until at least end of March.
Beech, J., & Chadwick, S. (Eds., 2004). The business of sport management. Harlow, England:
Prentice Hall/Financial Times.
Blake, A. (2005). The economic impact of the London 2012 Olympics, TTRI Discussion
Paper 5/2005, Christel De Haan Tourism and Travel Research Institute, Nottingham,
UK: University Business School.
Blake, A., Durbarry, R., Sinclair, M. T., & Sugiyarto, G. (2001). Modelling tourism and
travel using tourism satellite accounts and tourism policy and forecasting models. TTRI
Discussion Paper no. 2001/4.
Bourdieu, P. (1999). The state, economics and sport. In H. Dauncey & G. Hare (Eds.),
France and the 1998 World Cup: The national impact of a world sporting event (pp. 15–21).
London, UK: Frank Class.
Burns, J. P. A., Hatch, J. H., & Mules, T. J. (Eds., 1986). The Adelaide Grand Prix: The
impact of a special event. Adelaide, Australia: Centre for South Australian Economic
Studies.
Burrows, J., & Flynn, J. (2020, Apr 16). The impact of the Covid-19 Pandemic on sports
event cancellation insurance. https://www.sportresolutions.co.uk/news/view/the-impact-
of-the-covid-19-pandemic-on-sports-event-cancellation-insurance
Cherubini, S., & Iasevoli, G. (2005). Marketing for generating values in the event system.
In J. Andreani & U. Collesei (Eds.), Marketing trend in Europe. Paris: EAP Parigi.
Crompton, J. L. (1995). Economic impact analysis of sports facilities and events: Eleven
sources of misapplication, Journal of Sport Management, 9(1), 14–35.
Crompton, J. L. (2006). Economic Impact Studies: Instruments for political Shenanigans?
Journal of Travel Research, 45, 67–82.
Desbordes, M. (2007). The economic impact of a major sport event: The case of the
Tour de France among different regions, International Journal of Sport Management and
Marketing, 2(5-6), 526–540.
Desbordes, M. (2012). Football Event Management. In J. L. Chappelet & D. Aquilina
(Eds.), Handbook of football association management (pp. 117–144), UEFA Education
Programme. Nyon, Switzerland: UEFA.
Desbordes, M., & Falgoux, J. (2007). Les événements sportifs (3ème édition, préface de
Michel Platini). Paris: Editions d’Organisation.
Desbordes, M., & Richelieu, A. (2012). Global sport marketing: Contemporary issues and
practice, Routledge.
Dolles, H., & Söderman, S. (2008). Mega-sporting events in Asia – Impacts on society,
business and management: An introduction. Asian Business & Management, 7, 147–162.
Dolles, H., & Söderman, S., (Eds.) (2011). Sport as a business: International, professional and
commercial aspects. Houndmills, UK: Palgrave Macmillan.
Duncombe, W., & Wong, W. (1998). Building state and local government analytic
capacity: Using regional economic models for analysis of public policy. State & Local
Government Review, 30(3), 165–180.
Dwyer, L., & Forsyth, P. (2009). Public sector support for special events. Eastern Economic
Journal, 35, 481–499.
Dwyer, L., Mellor, R., Mistilis, N., & Mules, T. (2000). A framework for assessing ‘tangible’
and ‘intangible’ impacts of events and conventions. Event Management 6(3), 175–191.
256 L. Petrović, M. Ghasemzadeh, and C. Hautbois

Galvani, A., Lew, A., & Perez, M. S. (2020). COVID-19 is expanding global conscious-
ness and the sustainability of travel and tourism. Tourism Geographies, 22(3), 567–576.
Getz, D. (1994). Event tourism: Evaluating the impacts. In J. R. B. Ritchie & C. R.
Goeldner (Eds.), Travel, tourism and hospitality research: A handbook for managers and
researchers. New York, NY: John Willey & Sons.
Getz, D. (1997). Event management & event tourism. New York: Cognizant Communication
Corp-Broché.
Gratton, C., Dobson, N., & Shibli, S. (2000). The economic importance of major sporting
events: A case-study of six events, Managing Leisure, 5(1), 17–28.
Gratton, C., Shibli, S., & Coleman, R. (2006). The economic impact of major sports events:
A review of ten events in the UK. In J. Horne & W. Manzenreiter (Eds.), Sports mega
events: Social scientific analysis of a global phenomenon. Sydney, Australia: Blackwell.
Howard, D. R., & Crompton, J. L. (1995). Financing sport. Morgantown, WV: Fitness
Information Technology.
Humphreys, B., & Maxcy, J. (2007). The role of sport economics in the sport management
curriculum. Sport Management Review, 10, 177–189.
Jago, I., & Shaw, R. (1998). Special events: A conceptual and differential framework.
Festival Management and Event Tourism, 5(1/2), 21–32.
Jeanrenaud, C. (2006). Sports events: Uses and abuses of economic impact studies. Finance
& The Common Good/Bien Commun, 26, 99–103.
Johnson, A. T., & Sack, A. (1996). Assessing the value of sports facilities: The importance
of non-economic factors. Economic Development Quarterly, 10(4), 369–381.
Kaplan, R., & Norton, D. (1992). The balanced scorecard as a strategic management
system, Harvard Business Review, 1, 61–66.
Kaplan, R., & Norton, D. (1996a). The balanced scorecard – Translating strategy into action.
Boston, MA: Harvard Business School.
Kaplan, R., & Norton, D. (1996b). Linking the balanced scorecard to strategy. California
Management Review, 4, 53–79.
Kurscheidt, M. (2000). Strategic management and cost-benefit analysis of major sports
events: The use of sensitivity analyses shown for the case of the soccer World Cup 2006
in Germany. Neue Folge Nr. 69, Fachbereich Wirtschaftswissenschaften.
Leiber, S. R., & Alton, D. J. (1983). Visitor expenditures and the economic impact of public
recreation facilities in Illinois. In S. R. Leiber & D. R. Fesenmeier (Eds.), Recreation plan-
ning and management. State College, PA: Venture.
Matheson, V. A. (2002). Upon further review: An examination of sporting event eco-
nomic impact studies. The Sport Journal, 5(1). https://thesportjournal.org/article/a-
review-of-economic-impact-studies-on-sporting-events/
Matheson, V. A. (2006). Mega-events: The effect of the world’s biggest sporting events on local,
regional and national economies. Worcester, MA: College of the Holy Cross.
Matheson, V., Peeters, T., & Szymanski, S. (2012). If you host it, where will they come
from? Mega-events and tourism in South Africa. Antwerp, Belgium: University of
Antwerp.
Miller, A. (1998). Strategic management (3rd ed.). Irwin/McGraw-Hill.
Oldenboom, E. R. (2006). Costs and benefits of major sports events: A case study of EURO
2008. Amsterdam, the Netherlands: MeerWaarde Onderzoeksadvies.
Petrović, L. (2016). Strategic management and event management in light of generating compet-
itive advantages of organisational units: A resource-based approach. Belgrade, Serbia: The
University for Peace.
 COVID-19 and Economic Impacts 257

Petrović, L., & Desbordes, M. (2012, October). Globalisation and its aspects in the business
of sport. Paper presented at 18th International ECPD Conference on Reconciliation
and Human Security in the Balkans, Belgrade, Serbia.
Petrović, L., Milovanović, D., & Desbordes, M. (2015). Emerging Technologies and
Sports Events: Innovative Information and Communication Solutions, Sport, Business
and Management: An International Journal, 5(2), pp. 175–190.
Porter, P. (1999). Mega-Sporting Events as Municipal Investments: A critique of impact anal-
ysis. In J. Fizel, J. Gustafson, & G. Hadley (Eds.), Sport economics: Current research
(pp. 61–73). New York, NY: Praeger.
Preuss, H. (1999). Bedeutung des Verkaufs von Fernsehrechten für die Finanzierung sport-licher
Megaevents. In: H.-D. Horch, J. Heydel, and A. Sierau (Eds.), Professionalisierung im
Sportmanagement (pp. 258–277). Aachen, Germany: Meyer und Meyer.
Preuss, H. (2005). The economic impact of visitors at major multi-sport events. European
Sport Management Quarterly, 5(3), 283–305.
Preuss, H. (2010). Sports mega-events, sustainability & impact assessment: Summary report.
The University of British Columbia, Vancouver, BC: Centre for Sport and Sustainability.
Ritchie, J. R. B. (2000). Turning 16 days into 16 years through Olympic Legacies. Event
Management, 6(3), 155–166.
Rodríguez, P., Késenne, S., & García, J. (2006). Sports economics after fifty years: Essays in
honour of Simon Rottenberg. Ediciones de la Universidad de Oviedo.
Sandy, R., Sloane, P. J., & Rosentraub, M. S. (2004). The economics of sport: An interna-
tional perspective. New York, NY: Palgrave Macmillan.
Seraphin, H. (2020). COVID-19: An opportunity to review existing grounded theories in
event studies. Journal of Convention & Event Tourism, 20 (3-33).
Solberg, H. A., Andersson, T., & Shibli, S. (2002). An exploration of the direct economic
impacts from business travellers at the world championships, Event Management, 7(3),
151–164.
Solberg, H. A., & Preuss, H. (2007). Major sport events and long-term tourism impacts.
Journal of Sport Management, 21(2), 215–236.
Syme, G. J., Shaw, B. J., Fenton, D. M., & Mueller, W. S. (1989). The planning and evalua-
tion of hallmark events. Aldershot, UK: Avebury.
The Guardian (2020, October 13). England: Premier League aims for EFL bailout but some
clubs refuse to deal with Parry.
The Week (2020, September 22). UK: Football’s financial crisis: game is ‘unsustainable’ with-
out fans.
Trenberth, L. & Hassan, D. (2011). Managing sport business: An introduction. London:
Routledge.
Chapter 14

Brand Attributes Scale for

Professional Sport Teams
Measuring Strength, Favorability, and
Uniqueness of Team Attributes
Yukako Wada and Hirotaka Matsuoka

Introduction
Building a strong brand is one of the most critical issues not only for general
business companies but also for sport organizations. The value of brand assess-
ments, such as brand name, logo, and quality, is generally called “brand equity”
(Aaker, 1991; Keller, 1993). Since brand equity accumulates in consumers’ mem-
ories, customer-based brand equity (CBBE) is a significant concept of strategic
brand management (Keller, 1993). Brand knowledge, which is the core concept
of CBBE, consists of two aspects: brand awareness and brand image. A positive
brand image is created by strong, favorable, and unique brand associations. These
three types – strong, favorable, and unique – of brand associations are the key to
accumulating different brand equity in the minds of consumers. Therefore, the
concept and components of brand associations are crucial. Sub-dimensions of
brand associations and their scales have been examined and developed for assess-
ing corporate brands and services (Baalbaki & Guzman, 2016; Christodoulides
et al., 2009; Gordon et al., 2016; Grace & O’Cass, 2002; Low & Lamb, 2000;
Mann & Ghuman, 2014).
The concept of brand management has been a focus of the sport spectator
industry since the mid-1990s (Gladden, 2014). Researchers have recognized ath-
letes and sponsors for sporting events or sport teams as brands and examined con-
sumers’ perceptions and attitudes toward them (Boronczyk & Breuer, 2020; Girish
& Lee, 2019; Gwinner, 1997; Kunkel et al., 2020; Kwon & Shin, 2020; Taniyev &
Gordon, 2019; Won & Lee, 2020). Studies on sport teams have mainly discussed
brand association scales based on Keller’s (1993) CBBE concept because it assesses
structured brand equity that sport organizations can manage and control (Bauer
et al., 2005, 2008; Biscaia et al., 2013; Gladden & Funk, 2002; Kunkel et al., 2014,
2017; Ross et al., 2006, 2007, 2008).
The team association model (TAM; Gladden & Funk, 2002), based on Gladden
and Funk’s (2001) framework, was the first model to assess brand associations con-
cerning sport teams (Ross et al., 2007). Thereafter, Bauer et al. (2008) refined and
modified the refined TAM. The team brand association scale (TBAS; Ross et al.,
2006) was also developed and applied to measure brand associations concerning

DOI: 10.4324/9781003302476-14
 Brand Attributes Scale 259

collegiate and professional sport teams (Biscaia et al., 2013; Ross et al., 2007,
2008). Except for Bauer et al. (2008), these studies did not measure each of the
strong, favorable, and unique brand associations, even though these three types
are essential to brand success (Keller & Swaminathan, 2019). Bauer et al. (2008)
examined the relationship between fans’ team brand image and their loyalty by
measuring the strength, favorability, and uniqueness of associations regarding
professional sport teams. Their study, however, did not verify the validity of the
scale to measure associations representing teams.
The relationships between brand associations and team loyalty have been one
of the main topics, and researchers have also examined associations representing
sport teams based on fans’ perceptions over time as longitudinal research studies
(Bauer et al., 2005, 2008; Daniels et al., 2019; Gladden & Funk, 2001; Kunkel
et al., 2016; Martín et al., 2020; Wear & Heere, 2020). However, there is no aca-
demic scale to measure strong, favorable, and unique brand associations regarding
sport teams. Such a scale is also important for marketers to grasp their team brand
image based on consumers’ perceptions. Thus, it is crucial to develop a scale to
measure team brand equity in terms of their strong, favorable, and unique brand
associations.

Review of Literature
Keller (1993) proposed that brand associations are classified into three dimen-
sions: brand attributes, brand benefits, and brand attitudes. Brand attributes are
memories or thoughts that consumers recall when they buy or consume a par-
ticular product or service. According to CBBE, attributes are “descriptive features
that characterize a product or service” (p. 4). Brand attributes have two aspects,
namely product-related and non-product-related attributes. Product-related attrib-
utes are the “ingredients necessary for performing the product or service func-
tion sought by consumers”, while non-product-related attributes are the “external
aspects of the product or service that relate to its purchase or consumption”
(p. 4). Product-related and non-product-related attributes are categorized differ-
ently depending on consumers’ perceptions and needs.
Brand benefits are described as “the personal value consumers attach to the
product or service attributes” (Keller, 1993, p. 4). Attitudes refer to consumers’
overall evaluations of a specific brand that influence consumer behavior. Benefits
and attitudes are personal evaluations of attributes, and they are abstract asso-
ciations. Gladden and Funk (2001) did not measure attitudes because attitudes
are very abstract associations and formed through associations related to attrib-
utes and benefits. Individuals may recall attributes regarding a specific team first
and then evaluate them as benefits or attitudes. Nonpattern stadium attendees
who are not fans may find it difficult to recall benefits or attitudes toward a spe-
cific team. Therefore, this study develops the factors and measurement items of
team attributes that can measure strong, favorable, and unique brand associations
separately.
260 Y. Wada and H. Matsuoka

Brand Attributes in a Sport Team Context

Brand attributes are related to products or services and have two dimensions:
product-related and non-product-related attributes. Non-product-related attributes
include four sub-dimensions: price information, packaging or product appearance
information, user imagery, and usage imagery. The theory of “product levels”
(Kotler, 2003) also indicates five levels of products or services, and there are four
dimensions except for the core benefit that consumers require. Thus, non-product-
related attributes in the sport marketing context could be composed of more
than one dimension. The TAM and the TBAS, however, did not differentiate
between product-related and non-product-related attributes. The refined TAM
distinguished product-related from non-product-related attributes, but non-product-
related attributes had only one dimension.
The refined TAM mentioned that product-related attributes represented the core
product of a sport team (i.e., team performance), and non-product-related attributes
reflected external aspects of the core product (i.e., no relationship with team perfor-
mance). Therefore, this study defined brand attributes of the sport team as “descrip-
tive features of a sport team that help consumers to recognize its product or service”.
Product-related attributes were defined as “the elements of a certain sport team that
form the performance of games”. Non-product-related attributes were regarded as
“external aspects concerning the core product or service of a certain sport team
that do not affect the team performance of games directly”.
The refined TAM remodeled success, star player(s), head coach, and man-
agement of the TAM as five product-related attributes: success, star player, head
coach, team, and team performance. It also revised the logo design, stadium,
product delivery, and tradition of the TAM to four non-product-related attrib-
utes: logo and club colors; club history and tradition; stadium; and fans. With
regard to the 11 labeled groups of attributes of the TABS, non-player person-
nel, team success, team history, stadium community, team play, and brand mark
can be classified into product-related and non-product-related attributes of the
TAM. Three other TBAS attributes, namely rivalry, organizational attributes,
and concessions, which the TAM model lacks, can be categorized as non-product-
attributes. The other two labeled groups of attributes, namely commitment
and social interactions, can be classified as benefits (Kunkel et al., 2016). These
studies, however, focused on fans who show relatively high loyalty toward their
favorite teams. Mullin et al. (2014) divide stadium attendees into three segments:
frequent, multiple, and nonpattern attendees. Although the assessment of team
value based on fans’ perceptions is important (Kim et al., 2020), marketers also
need to understand the evaluations of not fans, such as occasional spectators
toward teams in their marketing for team management. Brand associations,
which are perceived by occasional spectators in the proper targeting markets of
teams, have not been studied.
The strong, favorable, and unique brand associations construct a differen-
tiated positive brand image that leads to brand success (Keller, 1993; Keller &
 Brand Attributes Scale 261

Swaminathan, 2019). It is important to assess the strength, favorability, and

uniqueness of brand associations based on consumer perception to understand
the characteristics of a certain brand. Keller (1993) indicated the concept of
strong, favorable, and unique brand associations, but the measurement items were
not indicated. The strength of brand associations can be measured by the number
or order of associative words and time when one recalls associations concerning
a certain brand association (French & Smith, 2013; Till et al., 2011). These are
based on objective methods, not consumers’ perceptions. According to the CBBE
context, strength shows the degree to which a consumer links a certain brand and
keeps a part of the brand image in one’s mind (Keller, 1993). As consumers receive
several stimulations by encountering and using a certain brand, their assessment
of a certain brand is based on their trace, which lasts a long time in their mind
(Brakus et al., 2009). Although consumers receive numerous information from
a certain brand, a part of it comes to mind when they recall the brand (Keller,
1993). Thus, the strength of brand association could be measured by consumers’
perceptions of whether a certain brand association always comes to their mind.
Favorability is the core concept of brand associations and consumers’ eval-
uation of how a specific brand association is favorable or not (Dacin & Smith,
1994). Favorable brand associations are pivotal for brand success because they
reflect the importance and benefits of a specific brand in one’s mind (Keller, 1993).
Consumers’ preferences, feelings, and beliefs toward an object are pivotal to form-
ing consumption behaviors (Blackwell et al., 2001). Although brand associations
do not always facilitate consumer behaviors, favorable associations lead to overall
positive attitudes toward a certain brand. Therefore, marketers need to under-
stand whether associations concerning professional sport teams are favorable or
not for their customers.
Uniqueness is key to creating a unique position within a market and pivotal for
maintaining competitive advantage (Keller, 1993; Southworth & Ha-Brookshire,
2016; Vieceli, 2011). Brand success requires unique firm associations that are
different from the others (Olsen, 2008). However, brand associations are often
shared with different brands, even though there are no competitors (Keller &
Swaminathan, 2019; Krishnan, 1996). It could be crucial that the associations
concerning professional sport teams are unique and identified as the associations
of a certain sport team. Thus, brand association concerning professional sport
teams must be perceived as unique by the customers.
Professional sport teams could be one of the symbols of their hometown. It
would be important to measure the strength, favorability, and uniqueness of team
attributes based on the inclusion of various insights, such as frequent, multiple,
and nonpattern attendees, to provide empirical implications for improving team
management and further develop the fan base. Therefore, the purpose of this
study was to develop a scale for measuring the strong, favorable, and unique brand
attributes of professional sport teams. As brand images of a sport team would be
both inside and outside of a stadium, brand associations regarding sport teams
should be collected by spectators who have attended a game at least once. This
262 Y. Wada and H. Matsuoka

study collects data from various types of sport consumers within proper targeting
markets of teams and composes brand attributes of professional sport teams with
product-related attributes and multiple dimensions of non-product-related attrib-
utes for scale development.

Method
The scale was developed in two steps: Step 1 involved collecting words represent-
ing professional sport teams from local residents where the teams were located
and then categorizing them to construct items for measuring team attributes. In
terms of non-product-related attributes, sub-dimensions of non-product-related
attributes were considered in the process of categorization. Step 2 included the
generation of items and determining the construct validity of the scale. The scale
comprised three models, namely the Strength, Favorability, and Uniqueness mod-
els. The Strength model measures the strength of team attributes, the Favorability
model measures the favorability of team attributes, and the Uniqueness model
measures the uniqueness of team attributes.
Data collection in Steps 1 and 2 were conducted in six cities where the home
stadiums of professional baseball teams were located in the Pacific League.
Professional baseball, operated by the Nippon (Japan) Professional Baseball
Organization (NPB), is one of the most popular spectator sports in Japan. The
NPB has 12 teams in two leagues – the Pacific League and the Central League –
each has six teams. Professional baseball teams have been considered advertising
tools of their owner companies in Japan. For the last two decades, however, the
business models of professional baseball teams, especially teams in the Pacific
League, have been developed following the European and American professional
sport team management. The marketing target of each team is local residents
where the home stadium is located, and teams have made efforts to strengthen
bonds with residents living in their marketing areas through promotional and
outreach activities for better team management (Wada & Matsuoka, 2016). Thus,
the Pacific League teams and their marketing targets (i.e., the local residents
where home stadiums of teams are located) may be valid subjects for developing a
scale to measure brand associations of professional sport teams.

Step 1

Data Collection
Data were collected from 930 respondents, 155 of which were from local residents
in six cities where the Pacific League teams are located. An online survey was
conducted through an internet research company. The sample included individu-
als over 16 years old who had registered with the Internet research company and
had attended at least one game between the 2012 and 2014 seasons. The question
used to collect words concerning team attributes was “Please write down up to
 Brand Attributes Scale 263

three words when thinking of ‘a specific professional baseball team’ in the order
they come to your mind”, we requested respondents to write down at least one
word. As the NPB baseball season runs from the end of March to the beginning
of November, the online survey was conducted in mid-February 2015 to avoid a
strong influence from game results and team successes or failures during or imme-
diately after the season.
A Ph.D. expert and five graduate students specializing in sport marketing dis-
cussed and categorized the words collected from local residents relying on the
TBAS, TAM labels, and refined TAM model. When the words were unclear and
unknown, the expert panel contacted the concerned teams. The entire process
was carried out based on the practitioner’s advice, who had worked for a Pacific
League team as a player and team staff member.

Findings
Among the respondents (N = 930), 56.67% were male, and 43.33% were female.
The mean age of the respondents was 47.50 years (SD = 13.69). Of these, 71.69%
had lived at their current residence for more than ten years, and 54.69% had
attended games one to three times in the last three years. The sex and mean
age of respondents were compared with the database of the Sasakawa Sports
Foundation (SSF) to confirm the representativeness of the sample. The SSF is
one of the most prominent Japanese research and survey institutions specializing
in sport. According to the 2014 database of the SSF concerning Japanese gen-
eral baseball spectators, 63.29% were male, and 36.71% were female. Their mean
age was 48.90 years (SD = 16.28). There was not much difference between the
respondents in Step 1 and the SSF database of 2014. Thus, the respondents in
Step 1 were recognized as a valid sample to represent the population.
A total of 1,967 valid words representing team attributes were obtained.
Categorization, labeling, and their definitions were considered while referring
to previous research studies (Bauer et al., 2008; Gladden & Funk, 2002; Ross
et al., 2006). The collected words were categorized into 15 labeled items: “team
performance”, “players”, “field manager”, “owner”, “team history”, “concessions
and attractions”, “team designs”, “mascot characters”, “relations with a commu-
nity”, “region”, “home stadium”, “fans”, “Pacific League and baseball”, “others”,
and “none”. Unmeasurable labeled items, “others” and “none”, and an item with
less than 5% of all associative words, “Pacific League and baseball”, were excluded
as measurement items. Consequently, 12 measurement items were composed to
measure attributes representing professional sport teams.
The 12 labeled and measurement items were divided into three sub-dimensions:
one product-related attribute (performance factor) and two types of non-product-
related attributes (internal factor and external factor). The product-related
attribute was related to the performance of teams in games. Therefore, it was named
the “performance factor”, and it included three measurement items, namely team
performance, players, and field manager. The remaining nine items corresponding
264 Y. Wada and H. Matsuoka

to non-product-related attributes were divided into the internal factor and the
external factor.
As mentioned previously, the CBBE has four sub-dimensions of non-product
attributes, namely price information, packaging or product appearance informa-
tion, user imagery, and usage imagery. However, the survey data featured no asso-
ciations regarding price information. Labeled items concerning team designs and
mascot characters are related to packaging or product appearance information
that Keller (1993) mentioned. Kotler and Armstrong (2011) argued that product
planners “need to develop product and service features, design, a quality level,
a brand name, and packaging” (p. 249). These are controllable factors by which
product planners deliver value to consumers. Every Pacific League team has a
parent company that hosts game day events and provides attractions using the
team’s history or parent company images. Team designs, mascot characters as
well as team history, owner, and concessions and attractions relate to quality
level, brand name, and packaging or product appearance information. Thus, the
internal factor comprised aspects that teams can manage directly at any location
or environment, namely team history, owner, concessions and attractions, team
designs, and mascot characters.
Relations with community, region, home stadium, and fans relate to user and
usage imagery described by Keller (1993). Therefore, the external factor was com-
posed of aspects that relations between teams and their external environments
that teams cannot manage directly, namely relationship with the community,
region, home stadium, and fans. Twelve labeled and measurement items and their
definitions are shown in Table 14.1. The structure of the team attribute scale is
shown in Figure 14.1.

Step 2

Item Generation
As 10 of the 12 labeled items, except region and home stadium, implied vari-
ous associative words concerning each label when the expert panel categorized
and labeled them, these items began with “regarding” as measurement items (e.g.,
regarding team performance or success; regarding active players). On the other
hand, region and home stadiums did not include their functions or services but
implied definite names as proper nouns. Therefore, the measurement items of the
region and home stadium were “a community or region where the home stadium
of the team is located” and “home stadium” (Table 14.1).
Each measurement item measured its strength, favorability, and uniqueness.
The question items to measure the strength of team attributes were, “When
thinking of a ‘specific professional baseball team’, how much do the following
items come to your mind?” This was measured on a 7-point Likert-type scale rang-
ing from “never” (1) to “always” (7). The question items to measure the favorabil-
ity of team attributes were, “when thinking of ‘specific professional baseball team’,
Table 14.1 Factors, Labeled Items, Definitions, and Measurement Items for Team Attributes

Factors Labeled Items Definitions Measurement Items

Performance factor Team performance Thoughts regarding the team performance or Regarding team performance or success
success of a specific team
Players Thoughts regarding the active players or their Regarding active players
performance and qualities in a specific team
Field manager Thoughts regarding the current field manager Regarding a current field manager
of a specific team
Internal factor Owner Thoughts regarding an owner or an owner company Regarding an owner or an owner company
of a specific team
Team history Thoughts regarding precursors and the history of Regarding the history of a team
a specific team
Concessions and Thoughts regarding concessions and attractions for Regarding concessions and attractions at
attractions a specific team at the stadium (except a game), such the stadium
as eating and drinking or special game day events
Team designs Thoughts regarding the symbolic designs, such Regarding symbolic designs of the team
as colors, logos, and slogans of a specific team (such as colors, logos, and slogans)
Mascot characters Thoughts regarding mascot characters of a specific Regarding mascot characters
team

Brand Attributes Scale

External factor Relationship with a Thoughts regarding the relationship between a Regarding the relationship between a team
community specific team and the community of the region and a community
where the team’s home stadium is located
Region The community or region where the home stadium A community or region where the home
of a specific team is located stadium of a team is located
Home stadium The home stadium of a specific team Home stadium
Fans Thoughts regarding supportive fans or fan cheers of Regarding fans of a team
a specific team

265
266 Y. Wada and H. Matsuoka

Figure 14.1 T
 he Structure of the Team Attributes Scale in the Strength, Favorable, and
Uniqueness Models.
 Brand Attributes Scale 267

how much are the following items seen as a good or bad image?” This was meas-
ured on a 7-point Likert-type scale ranging from “very bad image” (1) to “very
good image” (7). The question items to measure the uniqueness of team attributes
were, “when thinking of a ‘specific professional baseball team’, how much are
the following items unique?” This was measured on a 7-point Likert-type scale
ranging from “not unique at all” (1) to “very unique” (7). The scale validity was
confirmed with acceptable discriminant validity, convergent validity, and reliable
estimates in the Strength, Favorable, and Uniqueness models.

Data Collection
An online survey was conducted to confirm the construct validity of the scale
through an internet research company at the end of January 2016. Data were
collected from 171 to 173 residents in each city where the Pacific League teams
were located. The samples comprised respondents over 16 years old who were
expected to pay for themselves to attend games and had attended at least one
home game during the 2013 to 2015 seasons. The sample included 1,033 of the
1,034 responses (99.9%).

Findings
More than half (57.60%) of the respondents were male, and the female respond-
ents were 42.40%. The mean age of the respondents was 49.47 (SD = 13.51)
years. Among the respondents, 73.67% had lived at their current residence for
more than ten years. The sex and mean age of the respondents were compared
with the 2016 data of the SSF. Among Japanese general baseball spectators,
59.62% were male, and 40.38% were female. Their mean age was 48.34 (SD =
17.43). There was not much difference between the respondents in Step 2 and
those in the 2016 database of the SSF. They were quite similar to the character-
istics of the respondents in Step 1 and the database of the SSF in 2014. Thus,
the respondents in Step 2 were recognized as a valid sample to represent the
population.
The mean scores, standard deviations, and correlations across measurement
items of team attributes were calculated using SPSS Version 26 in terms of
their strength, favorability, uniqueness, and brand image as composite variables
(Table 14.2). The correlations of team performance and players were over .800 in
terms of the strength of team attributes. However, none of the items were elim-
inated because the correlations were lower than .700 for the favorability and
uniqueness of team attributes. Previous studies did not reject these items because
they were important for forming product-related attributes (Ross et al., 2007; Ross
et al., 2006).
Confirmatory factor analysis was conducted using Amos 26 to verify the valid-
ity of the scale. The chi-square to the degree of freedom ratio (CMIN/df), the
goodness of fit index (GFI), normed fit index (NFI), comparative fit index (CFI),
 268
Y. Wada and H. Matsuoka
Table 14.2 Means, Standard Deviations, and Correlation Ranges between Items for Strength, Favorability, and Uniqueness Models

Strength Model Favorability Model Uniqueness Model

Factors Labeled Items Mean (SD) Correlation Ranges Mean (SD) Correlation Ranges Mean (SD) Correlation Ranges

Performance Team performance 4.48 (1.82) .546**–.805** 4.66 (1.39) .397**–.583** 4.44 (1.19) .462**–.660**
factor Players 4.59 (1.79) .544**–.805** 5.22 (1.16) .583**–.693** 4.67 (1.11) .515**–.660**
Field manager 4.15 (1.91) .546**–.747** 4.98 (1.24) .511**–.633** 4.45 (1.23) .487**–.622**
Internal factor Owner 3.71 (1.83) .431**–.741** 4.50 (1.34) .479**–.716** 4.33 (1.23) .462**–.610**
Team history 3.68 (1.78) .542**–.733** 4.88 (1.15) 470**–.669** 4.58 (1.13) .514**–.644**
Concessions and attractions 3.74 (1.75) .445**–.695** 4.84 (1.20) .415**–.596** 4.45 (1.17) .422**–.556**
Team designs 4.03 (1.82) .446**–.703** 4.97 (1.18) .434**–.612** 4.58 (1.13) .489**–.602**
Mascot characters 4.01 (1.80) .507**–.705** 5.12 (1.21) .455**–.672** 4.70 (1.14) .422**–.630**
External factor Relationship with a 5.38 (1.86) .515**–.741** 5.66 (1.25) .474**–.716** 4.69 (1.23) .501**–.643**
community
Region 4.01 (1.77) .431**–.653** 5.06 (1.20) .416**–.671** 4.66 (1.18) .440**–.604**
Home stadium 4.89 (1.80) .546**–.706** 5.27 (1.24) .544**–.671** 4.81 (1.18) .496**–.644**
Fans 3.85 (1.90) .548**–.706** 5.18 (1.23) .397**–.681** 4.67 (1.19) .462**–.609**

Note: SD = standard deviation.

** p < .01.
 Brand Attributes Scale 269

and root mean square error of approximation (RMSEA) were used to evaluate the
goodness-of-fit criteria. The fit index of the Strength model was χ2/df = 3.34 (p <
.001), GFI = .984, NFI = .990, CFI = .993, and RMSEA = .048; the fit index of the
Favorability model was χ2/df = 3.09 (p < .001), GFI = .985, NFI = .989, CFI = .992,
and RMSEA = .045, and the fit index of the Uniqueness model was χ2/df = 2.86
(p < .001), GFI = .984, NFI = .986, CFI = .991, and RMSEA = .042. The goodness-
of-fit in each model was regarded as within the acceptable range.
Convergent validity was assessed via factor loadings (Table 14.3). The factor
loading that supports convergent validity is greater than .707 (Fornell & Larcker,
1981). Some factor loadings were lower than .707 (i.e., Team performance of the
Favorability model was .645; the Owner of the Favorability and Uniqueness mod-
els was .658 and .679, respectively, and the Region of the Strength model was
.650). None of the items, however, were excluded because the factor loadings for
other models (Strength, Favorability, or Uniqueness models) were greater than
.707. In terms of the average variance extracted (AVE), all items in all models
were greater than .50 (Hair et al., 2018). Discriminant validity was assessed by
comparing the squared correlation between two constructs with their respective
AVE (Table 14.4). The AVE for each construct must be greater than the squared

Table 14.3 Factor Loadings and Evaluation for the Goodness-of-fit of Models

Strength Favorability Uniqueness

Factors and Labeled Items Model Model Model

1. Performance factor
Team performance ←1 .863 .645 .784
Players ←1 .874 .917 .859
Field manager ←1 .855 .762 .724
2. Internal factor
Owner ←2 .801 .658 .679
Team history ←2 .845 .763 .737
Concessions and attractions ←2 .773 .743 .727
Team designs ←2 .874 .862 .805
Mascot characters ←2 .845 .806 .752
3. External factor
Relationship with a community ←3 .831 .791 .812
Region ←3 .650 .752 .725
Home stadium ←3 .818 .802 .797
Fans ←3 .767 .826 .757
Evaluation of the goodness-of-fit of models
The chi-square to the degree of freedom ratio 3.34*** 3.09*** 2.86***
(CMIN/df)
Goodness of fit index (GFI) .984 .985 .984
Normed fit index (NFI) .990 .989 .986
Comparative fit index (CFI) .993 .992 .991
Root mean square error of approximation (RMSEA) .048 .045 .042

*** p < .001.

270 Y. Wada and H. Matsuoka

Table 14.4 A
 verage Variance Extracted and the Squared Correlation between Two
Factors

Factors 1 2 3

Strength Model 1. Performance factor AVE = .75

2. Internal factor .63 AVE = .69
3. External factor .68 .67 AVE = .59
Favorability Model 1. Performance factor AVE = .61
2. Internal factor .60 AVE = .59
3. External factor .59 .61 AVE = .63
Uniqueness Model 1. Performance factor AVE = .63
2. Internal factor .60 AVE = .55
3. External factor .62 .63 AVE = .60

Note: AVE= Average variance extracted

correlation between that construct and any other constructs (Fornell & Larcker,
1981). The squared correlation between the internal and external factors was
greater than the AVE in the Favorability and Uniqueness models. All Cronbach’s
alpha and composite reliability (CR) scores were over .80. The reliability esti-
mates were also acceptable (Table 14.5).

Table 14.5 Reliability Estimates for Models

S F U

Factors Measurement Items α (CR) α (CR) α (CR)

Performance factor Regarding team performance .90 (.90) .80 (.82) .81 (.83)
or success
Regarding active players
Regarding a current field manager
Internal factor Regarding an owner or an owner .90 (.92) .89 (.88) .87 (.86)
company
Regarding the history of a team
Regarding concessions and
attractions at the stadium
Regarding symbolic designs of the
team (such as colors, logos, and
slogans)
Regarding mascot characters
External factor Regarding the relationship between .86 (.85) .88 (.87) .88 (.86)
a team and community
A community or region where the
home stadium of a team is located
Home stadium
Regarding fans of a team

Note: S = Strength model; F = Favorability model; U = Uniqueness model; α = Cronbach’s alpha;

CR = composite reliability.
 Brand Attributes Scale 271

Discussion
This study developed a brand attribute scale for professional sport teams
(BASPST) using the associations representing professional baseball teams based
on the CBBE. The BASPST was composed of three factors and 12 measurement
items, which considered both product-related and non-product-related attributes
and proposed more concrete dimensions for non-product-related attributes. The
items were used to measure strength, favorability, and uniqueness. The measure-
ment items were constructed based on the associative words representing teams
perceived by the local residents who lived in the teams’ actual targeted markets,
not only by fans. Furthermore, the validity of the BASPST was also confirmed
using data collected from local residents who had attended the home game at
least once within three years. Consequently, the BASPST was developed using
perceptions among various types of sport consumers’ perceptions within the tar-
geted markets of the teams. The scale validity was confirmed to be acceptable
in terms of the Strength, Favorability, and Uniqueness models. Therefore, we
concluded that the BASPST was valid for assessing brand attributes representing
professional sport teams.
This study had several limitations that require future research. The first is the
application of the BASPST to other sport teams. The scale development was
conducted by focusing on professional baseball teams and their marketing area.
Therefore, further examinations for scale validity should be confirmed by assess-
ing other types of sports teams so that the BASPST would enable researchers
and marketers to assess brand associations concerning professional sport teams in
different environments.
The second point is the discriminant validity of the scale, which is the same
concern as previous scale developments. Correlations between items were high,
and the squared correlation between the internal and external factors was greater
than the AVE in the Favorability and Uniqueness models. There are several pos-
sible reasons for these results. Promotions inside and outside the stadium and
outreach activities by professional baseball teams are becoming increasingly
diversified. Teams are becoming physically and psychologically closer to residents
through these activities and are blending into the daily lives of residents. The
more they blend in, the harder it is to distinguish among associations of profes-
sional sport teams. On the other hand, as the numerical values of the squared
correlation between the internal and external factors were greater than the AVE
(from only .02 to .08), revising the wording of measurement items may resolve the
problems concerning the discriminant validity of the scale.
The third point concerns brand benefits and attitudes. This study collected
words representing team attributes only and did not measure the strength, favora-
bility, and uniqueness of benefits or attitudes. However, in terms of benefits and
attitudes, whether individual benefits and attitudes regarding a particular product
and service are favorable or not could be a contradictory question. The refined
TAM (Bauer et al., 2008) did not measure the favorability of benefits because
272 Y. Wada and H. Matsuoka

they are intrinsically favorable associations. Attitude may also be a concept that
implies the perception of consumers as favorable or not regarding a particular
product or service. Therefore, in future research, the method for collecting words
or phrases regarding benefits and attitudes within marketing areas and measuring
their strength, favorability, and uniqueness should be considered when develop-
ing a scale.

References
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. New
York, NY: Free Press.
Baalbaki, S., & Guzman, F. (2016). A consumer-perceived consumer-based brand equity
scale. Journal of Brand Management, 23(3), 229–251.
Bauer, H. H., Sauer, N. E., & Exler, S. (2005). The loyalty of German soccer fans: Does a team’s
brand image matter? International Journal of Sports Marketing and Sponsorship, 7(1), 8–16.
Bauer, H. H., Sauer, N. E., & Exler, S. (2008). Brand image and fan loyalty in professional
team sport: A refined model and empirical assessment, Journal of Sport Management, 22,
205–226.
Bauer, H. H., Sauer, N. E., & Schmitt, P. (2005). Customer-based brand equity in the team
sport industry: operationalization and impact on the economic success of sports teams,
European Journal of Marketing, 39(5–6), 496–513.
Biscaia, R., Correia, A., Ross, S., Rosado, A., & Maroco, J. (2013). Spectator-based brand
equity in professional soccer. Sport Marketing Quarterly, 22, 20–32.
Blackwell, R. D., Miniard, P. W., & Engel, J. F. (2001). Consumer behavior (9th ed.). San
Diego, CA: Houghton Mifflin Harcourt.
Boronczyk, F., & Breuer, C. (2020). Brand-related feelings and sponsor attitude formation.
International Journal of Sports Marketing and Sponsorship, 21(3), 513–526.
Brakus, J. J., Schmitt, B. H., & Zarantonello, L. (2009). Brand experience: What is it?
How is it measured? Does it affect loyalty? Journal of Marketing, 73, 52–68.
Christodoulides, G., Michaelidou, N., & Li, C. H. (2009). Measuring perceived brand
luxury: An evaluation of the BLI scale. Journal of Brand Management, 16(5-6), 395–405.
Dacin, P. A., & Smith, D. C. (1994). The effect of brand portfolio characteristics on con-
sumer evaluations of brand extensions. Journal of Marketing Research, 31(2), 229–242.
Daniels, J., Kunkel, T., & Karg, A. (2019). New brands: Contextual differences and devel-
opment of brand associations over time. Journal of Sport Management, 33(2), 133–147.
Fornell, C., & Larcker, D. (1981). Evaluating structural equation models with unobserva-
ble variables and measurement error. Journal of Marketing Research, 18(1), 39–50.
French, A., & Smith, G. (2013). Measuring brand association strength: A consumer based
equity approach. European Journal of Marketing, 47(8), 1356–1367.
Girish, V. G., & Lee, C. (2019). The relationships of brand experience, sports event image
and loyalty: Case of Jeju International Ultramarathon Race. International Journal of
Sports Marketing and Sponsorship, 20(4), 567–581.
Gladden, J. (2014). Managing sport brands. In B. J. Mullin, S. Hardy, & W. A. Sutton
(Eds.), Sport marketing (4th ed., pp. 161–178). Champaign, IL: Human Kinetics.
Gladden, J. M., & Funk, D. C. (2001). Understanding brand loyalty in professional sport:
Examining the link between brand associations and brand loyalty. International Journal
of Sports Marketing and Sponsorship, 3(1), 54–81.
 Brand Attributes Scale 273

Gladden, J. M., & Funk, D. C. (2002). Developing an understanding of brand associations

in team sport: Empirical evidence from consumers of professional sport. Journal of Sport
Management, 16, 54–81.
Gordon, B. S., James, J. D., & Yoshida, M. (2016). The development of brand association
measures in multiple product categories: New findings and implications for goods and
service. International Journal of Business Administration, 7(3), 140–152.
Grace, D., & O’Cass, A. (2002). Brand associations: Looking through the eye of the
beholder. Qualitative Market Research, 5(2), 96–111.
Gwinner, K. (1997). A model of image creation and image transfer in event sponsorship.
International Marketing Review, 14(3), 145–158.
Hair, J. F., Black, W., Babin, B., Anderson, R. E., & Tatham, R. L. (2018). Multivariate data
analysis (8th ed.). Boston, MA: Cengage.
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand
equity. Journal of Marketing, 57, 1–22.
Keller, K. L., & Swaminathan, V. (2019). Strategic brand management: building, measuring,
and managing brand equity (5th ed.). Essex, United Kingdom: Pearson.
Kim, Y. D., Magnusen, M. J., Kim, Y., & Kim, M. (2020). Developing a sport fan equity
index. International Journal of Sports Marketing and Sponsorship, 21(2), 247–265.
Kotler, P. (2003). Marketing management (11th ed.). Upper Saddle River, NJ: Prentice Hall.
Kotler, P., & Armstrong, G. (2011). Principles of marketing (14th ed.). London, United
Kingdom: Pearson.
Krishnan, H. S. (1996). Characteristics of memory associations: A consumer-based brand
equity perspective. International Journal of Research in Marketing. 13(4), 389–405.
Kunkel, T., Biscaia, R., Arai, A., & Agyemang, K. (2020). The role of self-brand connec-
tion on the relationship between athlete brand image and fan outcomes. Journal of Sport
Management, 34(3), 201–216.
Kunkel, T., Doyle, J. P., Funk, D. C., Du, J., & McDonald, H. (2016). The development
and change of brand associations and their influence on team loyalty over time. Journal
of Sport Management, 30, 117–134.
Kunkel, T., Funk, D., & King, C. (2014). Developing a conceptual understanding of
consumer-based league brand association. Journal of Sport Management, 28, 49–67.
Kunkel, T., Funk, D., & Lock, D. (2017). The effect of league brand on the relationship
between the team brand and behavioral intentions: A formative approach examining
brand associations and brand relations. Journal of Sport Management, 31, 317–332.
Kwon, H., & Shin, J. E. (2020). Effects of brand exposure time duration and frequency
on image transfer in sport sponsorship. International Journal of Sports Marketing and
Sponsorship, 21(1), 170–190.
Low, G. S. & Lamb, C. W. (2000). The measurement and dimensionality of brand associ-
ations, Journal of Product & Brand Management, 9(6), 350–368.
Mann, B. J. S., & Ghuman, M. K. (2014). Scale development and validation for measur-
ing corporate brand associations. Journal of Brand Management, 21, 43–62.
Martín, F. V., Toledo, L. D., & Palos-Sanchez, P. (2020). How deep is your love? Brand
love analysis applied to football teams. International Journal of Sports Marketing and
Sponsorship, 21(4), 669–693.
Mullin, B. J., Hardy, S., & Sutton, W. A. (2014). Sport marketing (4th ed.). Champaign,
IL: Human Kinetics.
Olsen, E. L. (2008). The implications of platform sharing on brand value. Journal of
Product & Brand Management, 17(4), 244–253.
274 Y. Wada and H. Matsuoka

Ross, S. D., Bang, H., & Lee, S. (2007). Assessing brand associations for intercollegiate ice
hockey. Sport Marketing Quarterly, 16, 106–114.
Ross, S. D., James, J. D., & Vargas, P. (2006). Development of a scale to measure team
brand associations in professional sport. Journal of Sport Management, 20, 260–279.
Ross, S. D., Russell, K. C., & Bang, H. (2008). An empirical assessment of spectator-based
brand equity. Journal of Sport Management, 22, 322–337.
Sasakawa Sports Foundation. (2014). Sports life data 2014. Tokyo, Japan: Sasakawa Sports
Foundation.
Sasakawa Sports Foundation. (2016). Sports life data 2016, Tokyo, Japan: Sasakawa Sports
Foundation.
Southworth, S. S., & Ha-Brookshire, J. (2016). The impact of cultural authenticity on
brand uniqueness and willingness to try. Asia Pacific Journal of Marketing and Logistics,
28(4), 724–742.
Taniyev, O., & Gordon, B. S. (2019). Crafting a legacy: Investigating the retired athlete
brand image, International Journal of Sports Marketing and Sponsorship, 20(3), 390–406.
Till, B. D., Baack, D., & Waterman, B. (2011). Strategic brand association maps:
Developing brand insight. Journal of Product and Brand Management, 20(2), 92–100.
Vieceli, J. M. (2011). The measurement of the number, uniqueness and valence of brand
associations across three product categories. Journal of Customer Behavior, 10(3),
245–269.
Wada, Y., & Matsuoka, H. (2016). Community-based management: Developing the fan
base of professional baseball teams. International Journal of Sport and Health Science, 14,
31–40.
Wear, H., & Heere, B. (2020). Brand New: a longitudinal investigation of brand asso-
ciations s drivers of team identity among fans of a new sport team. Journal of Sport
Management, 34(5), 475–487.
Won, J., & Lee, J. L. (2020). The effectiveness of multiple brand portfolios: the role of
directional dominance and brand interference on brand evaluation. International Journal
of Sports Marketing and Sponsorship, 21(2), 285–303.
Chapter 15

Analyzing Athletes’ Satisfaction

toward the Organization
of a Sport Event
Development of a Questionnaire
Rosa Medina Rodríguez, Oswaldo Ceballos
Gurrola, Mireya Medina Villanueva, Marina
Reyes Robles, María Grethel Ramírez
Siqueiros, and Rosa López de D’Amico

Introduction
The National Universiade is the most important student sport event in Mexico,
as it gathers the best athletes from higher education public or private institutions.
Participating in them means the beginning of an elite sport career as it represents
a possibility to compete at the World Sport Universiade, Central American and
Caribbean Games, Pan American Games, and even Olympic Games. In these
events in which a big number of sport disciplines are involved, good management
implies the follow up of planning, development, and evaluation of the quality of
the organization. Nowadays, quality has become a need to guarantee the conti-
nuity and progress of the organizations, as it can generate benefits for all those
involved in the service process, such as clients (athlete), officials, employees,
among others, but overall, the image of the organization. Besides that, the world
of sport has been sensible to this issue and has incorporated it in its organizations,
as well as it has been of interest in the academic world (Calabuig & Crespo, 2009).
There are few studies that use instruments to evaluate participants’ satisfaction
of the organization of a massive sport event. So, through the level of satisfaction
obtained, key points can be identified in order to accomplish a successful event;
and with it, improvement aspects can be considered. The quality and satisfac-
tion of sport organizations and sport events have been the study reason of differ-
ent authors. In this sense, O’Neil et al. (1999) developed a scale for valuing the
service quality of a surfing sports event in Australia, based on the dimensions
of SERVQUAL reduced to three factors. Authors like Shonk and Chelladurai
(2008) have focused on valuing the satisfaction in a sport tourism event with four
factors: access quality, lodgings quality, and contest quality; in this same sector,
Kaplanidou and Gibson (2010) evaluated the relationships of service quality, sat-
isfaction, and future intentions of the attendees in a tourism sports event. They
DOI: 10.4324/9781003302476-15
276 R. Medina et al.

used the structural equations model, where the service quality is a significant
predictor of satisfaction. In another study made by Morquecho et al. (2016), the
QUNISPORT V.MX instrument was designed to evaluate the perceived qual-
ity in organizations of university sports services, with the aim of measuring five
factors: functionality and comfort of the university sport areas, interaction with
the coach, sport service management, characteristics of the sport activities pro-
gram, as well as environment and amenities of the bath/dressing rooms. Likewise,
Calabuig-Moreno et al. (2016) made the validation of a scale that measures spec-
tators’ perception of service quality in sport events through the structural equa-
tions model: the EVENTQUAL scale, which measures the spectators’ quality
in sport events: accessibility, personnel, tangibles, and complementary services.
MacIntosh and Parent (2017) carried out a study on athletes during a sports event
who answered the questionnaire about the experience of the athlete; the findings
showed a high level of satisfaction with the event. The real recipients and inter-
ested in the improvement of sports events are the athletes themselves, and for
this reason, their opinions must be analyzed, and it is essential to attend to their
needs (Bamford & Dehe, 2016; Hill & Hill, 2011). It is for this reason that the aim
of this study consisted in designing and validating a scale that allows valuing the
satisfaction with the organization and management of a massive sport event from
the athletes’ point of view.

Method

Participants
The population sample was of 706 athletes (364 women and 342 men), with an
age range of 17–26 years old (M = 21.04; DT = 1.86); they participated in a volun-
tary way, during the development of the University Games 2017 in Mexico, in 23
sport disciplines, both individual sports disciplines (chess, athletics, boxing, fenc-
ing, gymnastics, judo, karate, weightlifting, wrestling, Tae Kwon Do, tennis, table
tennis, archery, and triathlon) and team sports disciplines (basketball, baseball,
football, indoor football, handball, softball, volleyball, beach volleyball, and flag
football). The sample was divided randomly into two sub-samples. Sample 1 was
used to make the exploratory factor analysis (Study 1), made up of 354 athletes
(179 women and 175 men), with an average age of 20.83 years old (DT = 1.92).
Sample 2, which was used to make the confirmatory factor analysis (Study 2),
was integrated by 352 athletes (185 women and 167 men), with an average age of
21.26 years old (DT = 1.75). The age range and the type of sport discipline were
represented in a proportional way in both sub-samples.

Design of the Instrument

In order to design the Questionnaire of Athletes Satisfaction in the Organization
of a Sport Event (CUSADE, by its initials in Spanish), the one applied by Medina-
Rodríguez (2010) was taken as a reference, where 11 factors related with the
 Analyzing Athletes’ Satisfaction 277

satisfaction were evaluated, and it was developed in the Mexican sports context,
at municipal level. In this case, the level of satisfaction in aspects related with the
organization of a massive sport event, whose target population are the athletes,
for which the design of the present instrument, 10 factors with a total of 50 items
were designed: information (general information about the event), facilities, pro-
grammed activities, attitude (attitudes of the organizers, support personnel, ser-
vices), lodgings, transportation, food, medical service, security and vigilance, and
global valuing. A Likert-type scale was used to evaluate each item, with values from
1 to 5, where 1 is de minimum satisfaction and 5 is the maximum satisfaction.

Procedures
In order to design the CUSADE, a preliminary analysis was carried out. It con-
sisted of the revision of studies with a similar topic and the experience of experts
in sport management, where possible factors and items were discussed. They were
related with the satisfaction of the organization of a massive sport event. From
this point, criteria and pertinent instructions were established for the application
of the instrument. Thus, content validation was obtained. Then a meeting with
the responsibility of the National Sports Education Council (CONDDE, for its
initials in Spanish) and the university sports chairman was held in order to count
with the approval to carry out the research.
The application of the instrument was made during the development of the
university games, for which a group of trained survey takers visited the different
event areas (sport facilities, hotels, dining room, medical service, and the athletes’
rest area) through an explanation where they found out the aim of the study
and the way of completing the questionnaire, as well as any doubt or suggestion
related with it. The printed survey was self-administered, with the presence of the
survey taker. The average time of application was 15 minutes.

Data Analyses
In order to analyze the data, SPSS v.24, LISREL 8.80 (Jöreskog & Sörbom, 2006),
and the Mplus statistical programs were used. First, depuration and quality of
the database were carried out with the identification of the outliers; descriptive
parameters such as the mean and the standard deviation were obtained, as well
as the correlation between factors through the Pearson coefficient and Cronbach
alpha if the element is eliminated. An exploratory factor analysis was made, using
the KMO sample adequation measure and the Bartlett sphericity test; through
the extraction method of principal components and with the use of an oblique
rotation criterium (Promax, kappa = 4). Afterward, a confirmatory factor analy-
sis (CFA) attending the ordinal nature of the variables. Polychoric correlations
matrixes and asymptotic covariances were used as input, and the method of esti-
mation used was Maximum Verosimilitude (Bentler, 2006). Criterions of the χ2/gl
minor than 3.0 were considered in order to be considered a good adjustment of
the model (Kline, 2005). The CFI and NNFI indexes above 0.90 indicate an
278 R. Medina et al.

acceptable adjustment (Hu & Bentler, 1999). For the RMSEA, satisfactory val-
ues the ones minor than 0.05, and the inferior values are the ones bigger than
0.08 (Llorent-Segura et al. 2014). Once the CFA was concluded, standardized
coefficients were employed to determine the compound reliability indexes (CR)
and the extracted mean variance (AVE). Finally, the structural equations model
(SEM) was used to estimate causal relations chains, relating two or more latent
and manifested variables (Muthén & Muthén, 2007).

Results
The internal consistency was calculated through the Cronbach Alpha Index,
referring to the resulting factors that form the instrument. The reliability analysis
revealed a good internal consistency, showing values that, in this case, are greater
than 0.70 (Cronbach, 1951; Oviedo & Campos-Arias, 2005), among ranges of
0.83 and 0.92 (Table 15.1). The correlation analysis showed that all factors (infor-
mation, facilities, programmed activities, attitude, lodging, transport, food, med-
ical service, security, vigilance, and global valuing) correlate in a positive way
among them and are significantly related (p < .01).
In order to confirm the adequation of the instrument, an exploratory factor
analysis of the 50 items was made. The adequation of the sample was confirmed
through the Bartlett sphericity test and the Kaiser–Meyer–Olkin (KMO) adequa-
tion sample index. The value of the sample adequation measure was optimal,
with a KMO index of 0.94 and the Bartlett test resulted statistically with a value
(χ2 = 12769.4, gl = 1225; p < .001). Ten factors were extracted Eigen values of 1 or
greater than 1 which, altogether, explain a total variance of 70.95% (Table 15.2).
In order to confirm that the scale follows the expected factor structure, a confirm-
atory factor analysis was carried out through the LISREL 8.80 program (Jöreskog
& Sörbom, 2006). The adequation of the model was analyzed following some
recommendations through different indexes: the value of Chi-squared divided by
the degrees of freedom (χ2/gl) must be minor than three (Kline, 2015), the non-
normative adjustment index (NNFI), the comparative adjustment index (CFI)
above 0.90 (Hu & Bentler, 1995) and the average squared root of the proximity
error (RMSEA) inferior to 0.06 or a maximum value of 0.08 (Byrne, 2000). For
the previous, the indexes of the goodness of fit of the model resulted satisfactory:
χ2/gl = 2.55, NNFI = 0.96, CFI = 0.97, RMSEA = 0.07.
The reliability with internal consistency analysis, through the Cronbach Alpha
Index, revealed a good internal consistency, showing values in this case greater
than 0.70 (Cronbach, 1951; Oviedo & Campos-Arias, 2005) among a range of
0.84 and 0.92. The correlation analysis showed that all the factors are correlated
in a positive way among them and are significantly related p < .01. Moreover, a
calculation of the compound reliability coefficient (FC) based on confirmatory
factor analysis was made, taking as criteria the values bigger than 0.70 (Hair et al.,
2004, 2010). The result was with values between 0.87 and 0.92. The extracted
average variance (AVE) showed, likewise, adequate values, in this case superior, to
Table 15.1 Mean, Typical Deviation, Compound Reliability, and Correlations among the Variables

Factor Range M DT 1 2 3 4 5 6 7 8 9 10

1. Information 1–5 4.15 .68 (.88)

2. Facilities 1–5 4.33 .60 .532** (.83)
3. Programmed activities 1–5 4.42 .61 .547** .537** (.86)
4. Attitude 1–5 4.31 .72 .505** .561** .580** (.91)
5. Lodging 1–5 4.19 .80 .487** .382** .428** .483** (.89)
6. Transport 1–5 4.34 .72 .379** .437** .437** .633** .539** (.90)
7. Food 1–5 4.16 .73 .372** .283** .359** .446** .359** .516** (.87)

Analyzing Athletes’ Satisfaction 279

8. Medical Service 1–5 4.20 .85 .414** .349** .350** .390** .392** .423** .380** (.91)
9. Security 1–5 4.27 .79 .509** .505** .407** .540** .477** .547** .451** .548** (.92)
10. Global Valuing 1–5 4.40 .67 .531** .470** .475** .623** .538** .599** .468** .484** .657** (.89)

Note: FC = compound reliability; AVE = extracted mean variance; Cronbach’s alpha over the diagonal.
** p < .01.
Table 15.2 O
 rganization of the Rotated Factor Structure, Communalities, Self-Values, and Explained Variance Percentages (by Factor

280 R. Medina et al.

and Global)

Factor

Item Statement 1 2 3 4 5 6 7 8 9 10 Commonality

Ítem 2 The time it takes to serve you from the .821 .66
moment you request it
Ítem 4 The information received by any means of .740 .56
communication
Ítem 3 The information received at the time of arrival .740 .57
at the lodging site
Ítem 6 The information received upon arrival at the .721 .67
event site
Ítem 5 The attention you receive from the staff of .713 .66
the organizing committee
Ítem 1 Regarding the timetable of attention/ .675 .58
orientation
Ítem 7 How do you consider the personalized .668 .58
treatment received
Ítem 19 The organizing committee demonstrates a .886 .76
capacity to adapt to its interests
Ítem 20 The staff fulfilled their assigned functions .882 .78
Ítem 21 The staff was accessible to your .828 .76
requirements
Ítem 23 The staff was located in places where they .765 .74
could easily be reached
Ítem 22 Overall assessment of the attitude of the .697 .70
organizing committee
Ítem 15 The schedule of the games was respected .886 .72
Ítem 17 The general rules of participation were .845 .64
complied with. (Continued)
Table 15.2 O
 rganization of the Rotated Factor Structure, Communalities, Self-Values, and Explained Variance Percentages (by Factor
and Global) (Continued)

Factor

Item Statement 1 2 3 4 5 6 7 8 9 10 Commonality

Ítem 16 The referees were at the appointed time and .826 .69
place
Ítem 13 The games were held at the established time .707 .62
and place
Ítem 18 The games were held at the established .669 .58
place
Ítem 14 Overall evaluation of the programmed .554 .64
activities
Ítem 28 The stay at the assigned hotel was .885 .77
comfortable
Ítem 24 The distance between the hotel and the .868 .75
facilities was in accordance with the size of the city
Ítem 26 When you requested information you were .854 .79

Analyzing Athletes’ Satisfaction 281

attended in an adequate manner
Ítem 25 The organizing committee guided you .677 .68
regarding security measures in the hotel
Ítem 27 Overall rating of the lodging .647 .64
Ítem 36 The quality of the food received .915 .75
Ítem 37 The designated food service meets the .875 .76
expected needs and requirements
Ítem 33 The dining room and the staff in charge of it .795 .71
comply with health requirements
Ítem 35 The food area is in hygienic, clean and .685 .67
ventilated conditions
Ítem 34 Overall evaluation of the food service .631 .68
(Continued)
Table 15.2 O
 rganization of the Rotated Factor Structure, Communalities, Self-Values, and Explained Variance Percentages (by Factor

282 R. Medina et al.

and Global) (Continued)

Factor

Item Statement 1 2 3 4 5 6 7 8 9 10 Commonality

Ítem 38 There was a place and personnel marked as .914 .78
“Medical Service”
Ítem 39 The emergency personnel are trained to .842 .81
perform these functions
Ítem 41 The medical service designated for the event .809 .80
Ítem 40 Overall assessment of the medical service .795 .85
Ítem 45 Number of security personnel who guarded .924 .78
the event
Ítem 44 The attitude of the event’s security .896 .80
personnel
Ítem 43 How do you rate the security of the event? .892 .81
Ítem 42 Overall assessment of security and .794 .78
surveillance
Ítem 49 How do you rate your participation in the .818 .74
event?
Ítem 48 How satisfied are you with the organization .791 .76
of the event?
Ítem 50 If you were to participate in a similar event .773 .70
again, would you consider choosing the same venue?
Ítem 46 How would you rate your stay in the host .630 .73
city?
Ítem 47 How do you rate the attention you .581 .66
received?
Ítem 30 The transport is comfortable, clean and .874 .80
ventilated
(Continued)
Table 15.2 O
 rganization of the Rotated Factor Structure, Communalities, Self-Values, and Explained Variance Percentages (by Factor
and Global) (Continued)

Factor

Item Statement 1 2 3 4 5 6 7 8 9 10 Commonality

Ítem 29 The transport complies with the necessary .861 .78
safety conditions for its transportation
Ítem 31 The driver respects the current traffic .840 .79
regulations
Ítem 32 Overall evaluation of the transportation .791 .75
service
Ítem 8 Adequacy of the facility for sports activities .886 .68
Ítem 9 Cleaning and lighting of sports facilities .880 .73

Analyzing Athletes’ Satisfaction 283

Ítem 12 Signage and access to sports facilities .716 .66
Ítem 11 The attention received by the staff of the .563 .57
sports facilities
Ítem 10 Overall assessment of sports facilities .435 .62
Values Eigen 18.35 3.13 2.69 2.20 2.13 1.86 1.54 1.34 1.18 1.06 —
% of variance 36.71 6.27 5.39 4.40 4.27 3.71 3.08 2.68 2.35 2.12 70.95

Note: 1 = information, 2 = attitude, 3 = programmed activities, 4 = lodging, 5 = food, 6 = medical service, 7 = security y vigilance, 8 = global valuing, 9 = transport,
and 10 = facilities.
284 R. Medina et al.

0.50 (Bagozzi & Yi, 1988; Hair et al., 2004, 2010), in a range among 0.53 and 0.75
(Table 15.3). Analyzing the factors that form the instrument, five parameters of the
standardized solution were determined by conducting structural equation model
analyses, which are presented in Figure 15.1. For the analyses, the perception of the
received information and the programmed activities for the athletes was a predictor
in the satisfaction of the facilities and lodging; and this, at the same time, acts as a
positive predictor of the global valuing of the organization of a massive sport event.

Discussion
The aim of the present study consisted in designing and validating a scale that
allows valuing the satisfaction with the organization and management of a mas-
sive sport event from the athletes’ point of view. The results obtained in the pres-
ent study show adequate indexes for the scale, the internal consistency of each
of the resulting factors that conform it was excellent, showing that there exists
great homogeneity among the items of each factor, according to Nunnally (2008).
The confirmatory factor analysis applied shows that de indexes of the goodness
of fit of the model resulted satisfactory, which coincides with results obtained by
other authors that have used similar scales, such as Bernal (2013), Calabuing et al.
(2008), Kim and Trail (2010), and Medina-Rodríguez (2010). In this sense, the
CUSADEP showed a solid structure of ten factors.
Similar studies show scales with a smaller number of factors like the ones
from O’Neil, Getz, and Carlsen (1999), who evaluated three factors; Shonk and
Chelladurai (2008) four factors and Morquecho et al. (2016) included five fac-
tors to evaluate similar scales to one of the present studies. It is considered an
important aspect the fact of including wide range of factors, because it allows to
identify those that can condition the success of an event of this kind and, at the
same time, they allow the detection of negative effects on the satisfaction of the
clients/athletes (García et al., 2012). For this study, the degree of satisfaction of
the service respecting the information, facilities, programmed activities, attitude,
lodging, transport, food, medical service, security and vigilance, and global valu-
ing was evaluated positively for all factors, coinciding with other studies, in which
the totality of the scale factors was evaluated favorably (Bernal, 2013; Calabuing
et al., 2008; García et al., 2016; Medina-Rodríguez, 2010; Nuviala et al., 2008;
Ruíz-Alejos, 2015; Sánchez et al., 2017; Vila et al., 2009; Yildiz, 2011).
Rocha and Chelladurai (2012) don’t recommend the use of small samples to use
the design of structural models, for which counting with a big sample in the pres-
ent study allowed applying the structural equations model. This, at the same time,
leaded to identify the factors that best predict the satisfaction level of the partic-
ipants in a massive sport event, which correspond to those related with human
resources and tangible attributes, for which coincide with some authors (Bodet,
2006; Kim & Trail, 2010). They define them as determinants of the satisfaction
generation. In this sense, not only must the attributes with best valuing be taken
care of, but a bigger effort must be dedicated, for future sport events, to implement
Table 15.3 Mean, Typical Deviation, Compound Reliability, and Correlations among Variables

Factor M DE FC AVE 1 2 3 4 5 6 7 8 9 10

1. Information 4.15 .72 .89 .55 (.90)

2. Facilities 4.33 .65 .87 .53 .610* (.84)
3. Programmed activities 4.37 .68 .88 .55 .541* .557* (.88)
4. Attitude 4.30 .77 .91 .68 .512* .579* .531* (.91)
5. Lodging 4.17 .77 .87 .57 .551* .494* .427* .588* (.87)
6. Transport 4.27 .81 .92 .75 .414* .473* .444* .609* .515* (.92)
7. Food 4.05 .81 .90 .63 .447* .381* .412* .405* .458* .520* (.89)

Analyzing Athletes’ Satisfaction 285

8. Medical Service 4.13 .85 .91 .73 .491* .387* .370* .452* .415* .463* .427 (.91)
9. Security 4.27 .76 .91 .73 .505* .464* .424* .470* .326* .515* .402* .557* (.91)
10. Global Valuing 4.36 .68 .89 .61 .535* .597* .520* .667* .568* .618* .480* .509* .540* (.89)

Note: FC = Compound reliability; AVE = extracted mean variance; Cronbach’s alpha over the diagonal.
** p < .01.
286 R. Medina et al.

Figure 15.1 Standardized Solution of the Structural Model of the Satisfaction in the
Organization of a Massive Sport Event.

strategies in those factors that show a low-quality perception (García et al., 2012).
Understanding the service attributes is an important aspect for the researchers
and for the responsible of the enterprises in the sports sector (Albayrak & Carber,
2014). In this way, the results obtained in the present study are adequate to justify
and support the use of this scale and make it available for professionals as a tool
with adequate psychometric attributes, which will favor decision-making at the
strategic level (Gálvez et al., 2015).
The instrument represents a tool of easy application, and it has sufficient psy-
chometric guarantees in order to be used by researchers, professionals, and sport
managers for valuing the quality and satisfaction at the moment of organizing
massive sport events. Regarding the restrictions of this study, it is considered that
the specificity of the instrument itself can be considered a limitation; by virtue
of the fact that it can only be applied in massive events that involve all the factors
that define it and where the services and characteristics of the event are of such
magnitude. The model of structural equations allowed identifying the predictive
factors to accomplish global satisfaction, focusing on the received information by
the athlete attendees, the programming of activities of the event and those, at the
same time, related to the facilities where the event and lodging are being carried
out, identifying these four factors as the key elements that must be considered, at
the moment of organizing a sports event.

References
Albayrak, T. & Caber, M. (2014). Symmetric and asymmetric influences of service attrib-
utes: The case of fitness clubs. Managing Leisure, 19(5), 307–320.
Bagozzi, R. P. & Yi, Y. (1988). On the evaluation of structural equation models. Journal of
the Academy of Marketing Science, 16(1), 74–94.
 Analyzing Athletes’ Satisfaction 287

Bamford, D. & Dehe, B. (2016). Service quality at the London 2012 Games – A
Paralympics athletes survey. International Journal of Quality & Reliability Management,
33(2), 142–159.
Bentler, P. M. (2006). EQS 6 structural equation program manual. Encino, CA: Multivariate
Software.
Bernal, A. (2013). Fidelización de clientes en organizaciones deportivas: calidad, valor percibido
y satisfacción como factores determinantes (Tesis doctoral). Universidad Pablo Olavide,
Sevilla, España.
Bodet, A. (2006). Fidelización de clientes en organizaciones deportivas: calidad, valor percibido
y satisfacción como factores determinantes (Tesis doctoral). Universidad de Sevilla, Sevilla.
Byrne, B. M. (2000). Structural equation modeling with AMOS: Basicconcepts, applications,
and programming (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.
Calabuig, F. & Crespo, J. (2009). Uso del método Delphi para la elaboración de una medida
de la calidad percibida de los espectadores de eventos deportivos. Retos, 15, 21–25.
Calabuig, F., Quintanilla, I., & Mundina, J. (2008). La calidad percibida de los servicios
deportivos: diferencias según instalación, género, edad y tipo de usuario en servicios
náuticos. RICYDE. Revista Internacional de Ciencias del Deporte, 4(10), 25–43.
Calabuig-Moreno, F., Crespo-Hervas, J., Prado-Gasco, V., Mundina-Gómez, J.,
Valantine, I., & Stanislovaitis, A. (2016). Quality of sporting events: Validation of the
EVENTQUAL scale, Transformations in Business & Economics, 15, 2(38), 21–32.
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika,
16, 1–16.
Gálvez, P.; Boleto, A. & Romero, R. (2015). Validación de la versión reducida del
CECASDEP en usuarios de servicios deportivos. Suma Psicológica, 22(2), 78–85.
García, J.; Cepeda, G. & Martín, D. (2012). La satisfacción de clientes y su relación con la
percepción de calidad en Centro de Fitness: utilización de la escala CALIDFIT. Revista
de psicología del deporte, 21(2), 309–319.
García, J., Vegara, J., López, G., & Díaz, A. (2016). Satisfacción de usuarios de servicios
deportivos en Orihuela (Alicante). SPORT TK-Revista EuroAmericana de Ciencias del
Deporte, 5(1), 155–162.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (2004). Análisis multivariante.
Madrid, Spain: Pearson.
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis
(7th ed.). Upper Saddle River, NJ: Prentice-Hall.
Hill, A., & Hill, T. (2011). Essential operations management. London, UK: Palgrave
MacMillan.
Hu, L., & Bentler, P. M. (1995). Evaluating model fit. In R. H. Hoyle (Ed.), Structural
equation modeling. Concepts, issues and applications (pp. 76–99). Thousand Oaks, Ca:
Sage Publications.
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.
Jöreskog, K. & Sörbom, D. (2006). LISREL 8.80. Structural equation modeling with the
SIMPLIS command language (Software de cómputo). Chicago, IL: Scientific Software
International.
Kaplanidou, K., & Gibson, H. J. (2010). Predicting behavioral intentions of active event
sport tourists: The case of a small-scale recurring sports event, Journal of Sport & Tourism,
15(2), 163–179.
288 R. Medina et al.

Kim, M. & Trail, G. (2010). The effects of service provider employment status and ser-
vice quality exchange on perceived organizational image and purchase intention. Sport
Management Review, 13(3), 225–234.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New
York, NY: Guilford.
Lloret-Segura, S., Ferreres-Traver, A., Hernández-Baeza, A., & Tomás-Marco, I. (2014).
El análisis factorial exploratorio de los ítems: una guía práctica, revisada y actualizada.
Anales de psicología, 30(3), 1151–1169.
MacIntosh, E., & Parent, M. (2017). Athlete satisfaction with a major multi-sport event:
the importance of social and cultural aspects. International Journal of Event and Festival
Management, 8(2), 136–150.
Medina-Rodríguez, R. (2010). Opinión de los usuarios/deportistas acerca de los servicios
ofrecidos por una entidad deportiva municipal. Revista Mexicana de Investigación en
Cultura Física y Deporte, 2(2), 236–252.
Morquecho-Sánchez, R., Morales-Sánchez, V., Ceballos-Gurrola, O., & Medina-
Rodríguez, R. E. (2016). Cuestionario de evaluación de la calidad percibida en organ-
izaciones de servicios deportivos universitarios (QUNISPORT V.MX): programa factor.
Revista Iberoamericana de Psicología del Ejercicio y el Deporte, 11(2), 271–277.
Muthén, L., & Muthén, B. (2007). Statistical analysis with latent variables: User’s guide (6th
ed.). Estados Unidos: Muthén & Muthén.
Nunnally, J. C. (2008). Psychometric theory. New York, NY: McGraw Hill.
Nuviala, A., Tamayo, J., Iranzo, J., & Falcón, D. (2008). Diseño y validación de un instru-
mento de medida de la calidad de las organizaciones deportivas. Retos. Nuevas tendencias
en Educación Física, Deporte y Recreación, 14, 10–16.
O’Neil, M.; Getz, D. & Carlsen, J. (1999). Evaluation of service quality at events: The
Coca-Cola Masters Surfing event at Margaret River, Western Australia. Managing
Service Quality: An International Journal, 9(3), 158–166.
Oviedo, H. C., & Campos-Arias, A. (2005). Aproximación al uso del coeficiente alfa de
Cronbach. Revista colombiana de psiquiatría, 34(4), 572–580.
Rocha, C. M., & Chelladurai, P. (2012). Item parcels in structural equation modeling: An
applied study in sport management. International Journal of Psychology and Behavioral
Sciences, 2(1), 46–53.
Ruíz-Alejos, C. (2015). Análisis de la calidad percibida, satisfacción, valor percibido e inten-
ciones futuras de los usuarios de los servicios deportivos públicos gestionados por Logroño
Deporte (Tesis doctoral). Universidad de la Rioja, España.
Sánchez, C., González, C., López, G., & Díaz,A. (2017). Satisfacción de clientes externos.
Estudio de caso de una piscina cubierta. SPORT TK-Revista Euroamericana de Ciencias
del Deporte, 6(2), 81–88.
Shonk, D. & Chelladurai, P. (2008). Service quality, satisfaction, and intent to return in
event sport tourism. Journal of Sport Management, 22(5) 587–602.
Vila, I., Sánchez, C., & Manassero, M. A. (2009). Satisfacción percibida de los usuarios de
las instalaciones deportivas municipales de Palma de Mallorca. Revista Iberoamericana de
Psicología del Ejercicio y el Deporte, 4(1), 59–74.
Yildiz, S., & Kara, A. (2012). A re-examination and extension of measuring perceived
service quality in Physical Activity and Sports Centres (PSC): QSport-14 Scale.
International Journal of Sports Marketing & Sponsorship, 13(3), 27–50.
Index

Italicized and bold pages refer to figures and tables respectively, and page numbers
followed by “n” refer to notes.

Aaker, J. L. 207 analysis of variance (ANOVA) 193; model

Abeza, G. 146 error 42; one-way 10; two-way 10
ABI/Inform (Ovied) database 72 Analytical Facebook Data 151
absolute numbers 232, 232 analytic hierarchy process (AHP) 231,
acceptable fit 33, 44 231; decision-making theory 224; initial
accommodation 245 decision hierarchy 224–227; Saaty, T.L.
Achen, R. 145 224
actors 127 Anderson, J. C. 43
actual-self congruity 207 anecdotal information 101
Adamson, G. 146 Angela Stanford 119
adjacency matrices 114 annual reports 72
adoption of digital ticketing: athletic apparel 213
directors 95; perceived ease of use 95; Appel, G. 131
perceived usefulness 95; Technology Arefieva, V. 17
Acceptance Model (TAM) 95; see also Armstrong, G. 264
digital ticketing articles 72
advertising 131; strategy 127 Asada, A. 9
affiliations 97–99, 104 Asian Games at Hangzhou, 2022 218
after-sales service supply chain 220 asymptotically distribution free (ADF) 29
age 96 athleisure 126
Ajzen, I. 95 athlete endorsements 163
Akaike Information Criterion (AIC) athletes’ satisfaction 163, 243; data anal-
31–32 yses 277–278; design of the instrument
Alalwan, A. A. 132 276–277; discussion 284–286; elite sport
allegiance 201 career 275; factors 276; participants 276;
allegiant levels 211 procedures 277; results 278–284
allocative efficiency 248 athletic directors 101, 107
all-star game 172 attachment 201
Amazon Mechanical Turk (MTurk) 8, 9, attendance 97
43 attitude-like evaluative judgment 186
Ambush marketing campaigns 76, 88, 89 attitudes 96; and aptitude of people 70
American Sailing Cup 66 attraction 201
Amway 221 attractiveness 128–129, 138
Anagnostopoulos, C. 166, 167 attribute matrix 114
analysis data implementation 70 attrition 7
290 Index

audio-visual technology 55 Blair, M. M. 222

augmented reality (AR) 56; sport sector Blizzard’s Overwatch League 63
61–63 blogs 65
Australian football team: rules 6 Bollen, K. A. 30, 43
Australian Rules Football 145 Bonacich, P. 118
authenticity 138 Bonacich’s Beta Centrality 118, 120
auto-ethnography 150, 151 Bonferroni procedure 10
average variance extracted (AVE) 13, 44, Bonsón, E. 145, 157
45, 269–271, 270, 278, 285 books 213
awareness 201, 211, 212 bootstrapping 37, 38, 42
Aydin, G. 153, 155 bottom-up approach 246
box office 105
Baby Boomers 96 Brand Alliance Concept 128
“badness of fit” index 30 brand associations 205, 260–261
Baena, V. 75, 77, 82, 88 brand attributes 165–167; brand associa-
Bagchi, A. 75, 79, 81, 85 tions 258; consumers’ perceptions and
Bagozzi, R. P. 43 attitudes 258; data collection 262–263,
balanced scorecard (BSC) approach 258; 267; discussion 271–272; findings
customer/socio-cultural impact perspec- 263–264, 267–270; item generation
tive 253; financial/economic impact 264, 267; sport team context 260–262;
perspective 252; internal business pro- sub-dimensions 258; types 258
cesses/service impact perspective 253; brand attribute scale for professional sport
learning and growth 253–254; stake- teams (BASPST) 271
holders’ approach 252, 253; strategic and brand category 202
financial controls 250 brand equity 258
Ballings, M. 145, 155, 156, 158, 159 brand image 71
Barbaranelli, C. 207 brand loyalty 51
Barcelona FC 60 brand promotion 163
Baron, R. M. 34, 36, 37 brands growth 52
Bartlett sphericity test 277 Brett, J. M. 27, 28, 36–38
Basketball 145 broadcaster to viewer 55
basking in reflected glory (BIRG) 18, 152 Burton, N. 75, 77, 82, 88, 89
Bauer, H. H. 166, 259 business intelligence (BI): application 17
Bayesian Information Criterion (BIC) 32 business strategies 251
Bay to Breakers in San Francisco 212, 213 business-to-business (B2B) companies 132,
Bearden, W. O. 184 145
Beaton, A. 201, 205, 206, 211, 214 business-to-consumer: businesses 145;
Beccarini, C. 185 companies 132
Bee, C. 153 Byon, K. K. 5, 13–15, 26, 42
behind-the-scenes 112 Byrne, B. M. 42
Beijing Olympics, 2008 218
Beijing Winter Olympics, 2022 218 Cabrera-Sánchez, J. P. 18
beliefs 96, 261 Calabuig-Moreno, F. 276
Bentler, P. M. 30, 31, 44 Call of Duty 63
Bernal, A. 284 Calvo-Porral, C. 136
Bernstein, I. H. 208 camaraderie 182
best-fitting model 32 Caprara, G. 207
beta centrality scores 117 Carlsen, J. 284
betting 56 Caro, L. M. 185, 186
bi-directionality 27 case studies 3; defined 4; exploratory
big data: components 18; data-analytic research 4
techniques 18 cash walk-up purchase 94
big databases 66 casuals 247
 Index 291

categorization 263 comparative fit index (CFI) 31, 267

causal inference (CI) 5 competing models 13, 35, 45
causal relationship 251; causal order competition of product attributes 188
specification 27; empirical confirma- complementary assets 52
tion 28; empirical disconfirmation composite reliability (CR) 270
28; operationalization of variables 27; compound reliability indexes 278, 279,
“priori” 27; self-containment 27; speci- 284, 285
fication of boundaries 27; specification computable-general-equilibrium (CGE)
of the causal direction 27; stability of modeling 244, 249–250
the model 27; structural model, formal computer-generated digital content 62
statement of theory 27; theoretical conceptualizations 203
rationale of causal hypotheses 27 conditional index (CI) 173
cause-and-effect: chains 251; relationships conference affiliation 81
7, 8, 252 configural invariance 40
cause-related marketing (CRM) 10 configural model 42
celebrities 114, 116 confirmatory factor analysis (CFA) 11–12,
celebrity endorsement: accusations 128; 277, 284; factor structure 12; maximum
advertising strategy 127; controversy likelihood estimation 13; WLS estima-
128; definition 127; financial costs 128; tion method 29
ill-behavior 128; language, culture, and congruency: data analyses 208; functional
time zones 128; models and theories congruity 202; measures 205–208; par-
128–130; money 128; televisions ticipants 204–205; types of 202
advertisements 127 congruity: brand association 205;
Central American and Caribbean Games functional 210; stereotypical user 205;
(CAC/CACGs) 275 symbolic 210
centrality 205 connected stadium 60
Chanavat, N. 89 Consistency Index (CI) 233
Chelladurai, P. 275, 284 Consistency Ratio (CR) 233, 234
Chen, B. 224 Consolidated Standards of Reporting
Chen, H. G. 132 Trials (CONSORT) checklist 72
Cheon, J. 155, 160 constant connectivity 56
Chicago Marathon 212 constrained model 42
Chinese Professional Baseball League constraint-effects-mitigation model 13–4
(CPBL) 164 construct reliability (CR) 33, 44
Chinese Super League soccer tournament construct validity 44
223 consumers 55; assessment 183; behavioral
chi-square: difference test 33, 34, 45; test intentions 71, 186; characteristics 188;
value 30, 31 commitment 201; congruency 202–204;
Cho, C. 155, 160 decision-making process 5, 203;
Cho, D. W. 220 definitions 203; demand satisfaction
Chun, C. S. 137 224, 231; demographic characteristics
Chung, K. 9, 10 95; demographics 93; experience 88;
Cianfrone, B. A. 97 intention 165; internal reference 183;
coach 276 memories 258; perceptions 103, 259, 261;
cohesive subgroups 114 preferences 261; purchase 133; purchase
collaborative work 70 intention 128, 137
common method bias (CMB) 5 consumption traits: community-driven 3;
communalities 278, 280–283 emotional 3; making their voices heard
communication: channel 16; effective 2; 3; millennial consumption 4; peer pres-
and interactivity 9; and marketing deals sure influenced 3; technology driven 3
65 contextual interactions 71
community-wide activities 187 contract fulfillment 224, 231
comparative adjustment index 278 convenience fees 100
292 Index

convergent validity 44, 269 data-based marketing (DBM) 2

coordination 231 data collection 246, 262–263; analysis
corrected item-total correlation (CITC) method 127
228–230 data collection methods: cross-sectional
correlations 271, 278, 279, 284, 285; study 5; longitudinal study 5
analysis 32, 278; ranges 267, 268 data collection strategies 4; collecting data
Cortina, J. M. 38, 39 5; employing multiple respondents 5;
cost-benefit analysis (CBA) 244; benefits gathering, data 5
and costs 248; economic impact anal- data extraction and synthesis: analyses
yses 247; qualitative and quantitative performed 73; country 73; data collec-
assessment 248; strategic management tion 73; flow diagram 72, 73; journal
248; well-defined object of examination title 73; method 73; platform 73; results
248 and application 73; sport 73; test quality
cost estimations 252 73–74; theory 73; type of organisation
covariance: differences 31; matrix 28 73; year of publication 73
covariance-based SEM (CB-SEM): Davis, F. D. 96
implied-covariance matrix 14; structural day-to-day operations 219
model handling 14 decision-makers 96, 227
covariant relationship 27 decision-making: consultants 227; process
COVID-19 pandemic 90, 93; balanced 53, 133, 136, 137, 203
scorecard approach (BSC) 250–254; degree of satisfaction 284
computable-general-equilibrium (CGE) Deloitte’s digital media 57
modeling 249–250; cost-benefit analysis DeMers, J. 155
(CBA) 247–249; global health crisis demographic influences: affiliation 97–98;
243; input-output analysis 244–247; age 96; education 96–97; gender 96;
lockdowns 243; public and not-for-profit income 97; party composition 98; race
organizations 258; sport as business 97
258; sports management literature 258; Desbordes, M. 89
tourism and cognate industries 243 descriptive research design 4–6; sport mar-
credibility 39, 138; expertise 128; keting stakeholders 4; types 5
trustworthiness 128 descriptive statistics 32, 208, 209
Creswell, J. 149 De Vries, L. 178
Cronbach Alpha Index 278 Dick, R. J. 75, 77, 81, 82
Cronbach’s alphas 33, 45; coefficients 208; Dickson, G. W. 220
tests 189; values 44 diffusion 7
cross-cultural studies 5 digital communication 54, 59, 88; typolo-
cross-legged panel model 5 gies 59–63; see also digital marketing
cross-sectional study 5 digital competitors 71
crowd control 94 digital economy 56
cumulative satisfaction approach 185 digitalization 65
CUSADE 276, 277, 284 digital marketing: academic variables
customer: loyalty 52; satisfaction 185, 186 of the articles 74, 75; applications
customer-based brand equity (CBBE) 258, of the studies 81, 82–87; attracting
259, 261, 264, 271 customers 71; business models 70;
customer relationship marketing (CRM) clients 71; COVID-19 pandemic 90;
16 data extraction and synthesis 72–74;
cut-off value 30, 31, 44 hashtag 89; inclusion and exclusion
cutting off reflected failure (CORF) 18, criteria 72; internet 89; limitations
153 89–90; methodological aspects of the
Cvijikj, I. P. 177 articles 76, 77–80; methodological
quality assessment 72; mixed meth-
Dahnil, M. I. 132 ods 76; non-sport-related journals 74;
data analyses 208 organization attention 71; product
 Index 293

promotion campaigns 88; six descriptive EdgeRank algorithm 155

studies 81; social media 71; techniques education 96–97; intentions 100, 106
and strategies 81; transformation 70; Edwards, J. R. 38
website, email, apps 71; website or social Eigenvalue 12
networks 76 eight-participation-based running events
digital technologies 56; BI 17–18; big data 146
17–18; broadcasters 65; co-creation pro- eight-step procedure 39
cess 67; smart clothing 60; social media Eisenhardt, K.M. 4
analytics 16–17; sport distribution 66; electronic equipment 53
sport event content 58; sports buildings electronic signage 60
60; web analytics 16 electronic word of mouth (eWOM) 133,
digital ticketing: athletic directors 93; 134
companies 94; crowd control 94; demo- Eleven Groups Users 116, 117
graphic influences 95–98; education e-marketing see digital marketing
106; ethnicity 106; factors influencing empirical confirmation: of the model
adoption 95; fans and spectators 95; proposed model 28
high school athletic events 94; level empirical disconfirmation 28
of income 106; managerial decisions empirical reasoning 33
93; marketing campaigns 94; market employment opportunities 245
segmentation 95–98; method 98–99; endogenous: latent variables 43; variables
results 99–100; revenue 93; school-as- 26
signed budgets 93; spectators’ online endurance event: consumer behavior
purchases 93; sponsorship packages 94; research 199; consumer congruency
sport administrators and marketers 93; 202–204; data analyses 208; demo-
technology-based feature 93; ticketing graphic profile 205; involvement and
cash box 94 congruity 210–211; limitations and
digital website marketing 89 future research 213–214; marketing
direct effects 245 communication channels 199; measures
direct marketing 131 205–208; non-traditional events 199;
disagreements 72 participants 204–205; performance
discriminant validity 271; interfactor brands 199; practical implications
correlations 44 212–213; psychological continuum
displays 53; monitor 60 model (PCM) 201–202; results 208–210;
disruptive technologies: digital acquisition self-congruity 200; symbolic meaning
era 57; network of collaborative rela- 199–200; theoretical implications
tionships 51; sport clubs 51; sport event 211–212; utilitarian features 199
marketing mix 53, 54; sport events 51; English football pyramid 243
sporting equipment suppliers 52 English Premier League (EPL) fans 16
distribution routine 62 entertainment expenditures 245
distributive justice 183, 184 Entertainment & Sports Programming
Djafarova, E. 136, 137, 138 Network (ESPN) 62
dual-channel model 14 entrepreneurship 1
dual entitlement theory 183, 184 Environment 3.0 90
Duncan’s post hoc test 192 environmental factors 188
dynamic sport industry 188 environmental pollution 247
equipment 213
Eagleman, A. 152 equity theory 183
Ebrahim, R. S. 132 equivalence of factor: covariances 41;
ECO certification 221 variances 41
E-commerce 221 equivalent models 34
econometric approach 246 Erdogan, B. Z. 128
economy 252 eSports 52, 53, 56; demands 63;
Eddy, T. 16 demographic 57; disruptive
294 Index

affirmation 57; division 63; financial risk 149–151; engagement 147–148; insights
58; inclusive and virtually immersive 57; 150, 150, 151; likes 148; page of conven-
living spaces for teams 58; market 63; ience stores 165; page reach 148; photos
marketers 56; potentials 58; revenues and videos 155; popularity 144; reach,
56, 57; sponsorships 56; tournaments engagement 148; real-time direct com-
63; traditional team sports 58; training munication 147; and shares 148; social
facilities 58; venues 58; “winning” and media 144; total page engagement 148
“losing” concepts 59 Facebook performance: engagement 156;
estimation: method types 29; procedure 33 posts 156–157; reach 155–156
ethnography 150 Facebook proportional performance:
European and American professional sport engagement 158; posts 158–160; reach
team management 262 157–158
evaluation reports 72 faceto-face relationships 70
Evans, D. M. 75, 77, 81, 83 face value 151
event-oriented sport 236 facility aesthetics 53
EVENTQUAL scale 276 factors: loadings 44; variance equivalence
event-related costs 248 41
event-related expenditures 247 Fader, P. S. 188
exclusion criteria: articles in a language familiarity 128
other than English, 72; Congress pro- fan-created Facebook pages 164
ceedings, book chapters, books, or other fan-created social media 164
types of publications 72; duplicate arti- fans 163; attachment and loyalty 53;
cles 72; no digital marketing actions 72; awareness 54; championship games
theoretical studies, qualitative approach 58; community 65; connection, and
or reviews 72 attachment 146; eSport 56; hardcore 60;
exogenous variables 26 live participation to 60; “meme-ified”
expectation maximization (EM) 42 pictures 58; photographed pictures 58;
expected cross-validation index (ECVI) 34 and supporters 52, 60; virtual
experience interviews 3 participation 60
experiential goods/services 185–186 fan satisfaction index (FANSAT) 14
experimental research design 6–9; fans’ co-creation behaviors 165
cause-andeffect relationships 7; external Fantasy League Organizers 63
validity 7–8; internal validity 7; qua- fashion 126
si-experimental design 7; true design 7 Faulds, D. J. 131
explained variance percentages 278, favorability 261, 262, 271
280–283 feelings 261
explanatory variables 169, 169 female professional golfers 117
exploratory design 3–4 Ferrand, A. 185
exploratory factor analysis (EFA) 11–12; FIFA World Cup 18, 246; South Africa,
factor extraction methods 12 2010 249
external validity 7–8; field experiments 8; Filo, K. 71, 73–75, 78, 84, 88, 89
psychological realism 8; replicating the financial databases 76
study 8; statistical method 8 finishers 211
first-generation techniques 11
Facebook 15, 16, 76, 112, 130; admin- first-time finishers 211
istration 154–160; analytical pages first-time participants 213
150; baseball teams 164; benefits 146, Fishbein, M. 95
152–153; brand pages 147–149; case fit indices 34
context 149; challenges 153–154; fitness 126; apparel 126; equipment
comments 148; components of engage- 52; industry 126; marketing strategy
ment 147; content and performance 145; 126; micro-celebrities 135–136; and
data analyses 151–152; data collection recreation-focused sport 236
 Index 295

fitness-focused users 135–136 women golfers 110; Korea’s economy

fitness-for-all 218, 236 110; Ku, Ok-Hi 110; Pak, Se-Ri 109–110;
fitness switching costs model (FSCM) 14 sponsoring 111
fit statistics 35, 36 golf-tournaments 114, 116; users 117, 121,
Five-Factor Proposed Model 26 122
The 13th Five-Year Plan of Sport goodness-of-fit 30, 31; indexes 45; of
Development 218 models 269, 269; structural model 46
The 13th Five-Year Plan of the Sport Industry goodness of fit index (GFI) 267
218 Google Analytics (GA) 16, 98
flagship 61 Google Cardboard 61
focus group 4 Google search engine 15, 16
Football World Cup 66 Google Trends (GT) 16
foreign market of choice 128 government reports and documents 3
Fornell, C. 44 Govers, P. C. 207
The Four Ps 53 Great Recession in 2018 1
Fox Sports 62 Greenwell, T. C. 183, 184, 189
Freeman, R. E. 222 Grey literature 72
Fubon Guardians’ group 169 Gross domestic product (GDP) 244
full-text review 72 gross value added or “net output” (GVA)
functional attributes 201 245
functional congruity 205, 209, 210 group differences 42
Funk, D. C. 71–75, 78, 88, 166, 205, Guangzhou Asian Games, 2010
258 218
fun or enjoyment 202 Guido, G. 207
Future Internet (journals) 74 Gyulai, D. 165

gamification 57 Ha, I. 96
gaming 57–58 Hair, J. F. 13, 31, 43, 44
Garcia, J. A. M. 185, 186 half-marathons 205, 211
GASC 218, 223 Handfield, R. B. 219–220
gender-based discrimination 96, 112 Hawkins, D. I. 97
generalized least squares (GLS) 29 Haws, K. L. 184
General Motors 221 Hazari, S. 75, 78, 84, 89
Generations X 57, 96; sport fans 42 head coach 166, 260
Generations Y 96 Hendry, J. 222
Gerbing, D. W. 43 Henkel 221
Getz, D. 284 hide post 155, 160
Gibson, H. J. 275 hierarchical regression analysis 174–176,
Gillan, P. 152 184, 191, 191
Gillooly, L. 89 hierarchy of strategic intent 248
Gladden, J. M. 166, 258 high-intensity competition 67
glass-walled rooms 60 high school representative 99
global audience 58, 67 Hongkou Football Stadium 223
globalization 113 Hongshuang, A. L. 70
global performers teams 66 hosting 246
global sales 126 Howat, G. 186
global valuing 284 Huang, W. 224
GOFAN 94 Huddle Inc. (ticketing company) 94
golf associations 114, 116 Hu, L. 30, 31, 44
golf courses/clubs 114, 116 Hwang, G. 9, 10
golf industry in South Korea: golf-related Hyatt, C. G. 4
purchases 110; KLPGA 110–111; Korean hyper-connected habitué 56
296 Index

hypotheses 42, 43, 165; conceptual internal validity: association without

framework 167, 167; structural model 45; intervention 7; blinding the partici-
testing 41 pants 7; random assignment 7; random
selection of participants 7; rigorous
implied-covariance matrix 30 study protocol 7
inclusion criteria: digital marketing in a International Champions Cup (ICC)
sport context 72; journal articles 72; soccer 61
online advertising 72; publications in International Journal of Sport Management
English 72 11
independent variables 24 International Journal of Sports Marketing &
index weight assessment 234–235 Sponsorship 74
index weight value analysis: consistency internet market see digital marketing
analysis of interval judgment matrix interscholastic sports 107
233–234; decision hierarchy structure interval judgment matrix 233–234
231; determine the index weight 234, invariance: configural 40; confirmatory
234; index weight assessment 234–235; factor analysis (CFA) 45, 46; equiva-
item grading 234–235; pairwise com- lence of factor 41; item uniqueness 41;
parison matrices 232–233; reranking metric 40; scalar 40; structural equation
234–235 modeling (SEM) 42
indicator loadings 44 investors 65
indirect effects 245 involvement levels 205, 210–213
indoor-leading sport facilities 237 Ioakimidis, M. 75, 78, 81, 84, 88
induced effects 245 Ironman event 202
information and communication technol- Ironman triathlon series 211
ogy (ICT) 16; evolution 55 Itani, O.S. 184
information table 98 item generation: home stadium 264;
initial decision hierarchy: outline 224–226; Likert-type scale 264, 267; unique 267
refinement 224, 227; sport facilities ser- item grading 234–235
vice provider selection 224, 225–226 item uniqueness 41
Inoue, Y. 3
input-output (I-O) analysis 244; economic Jago, I. 244
impact 245; inter-industry linkages 244; James, L. R. 27, 28, 36–38, 205
major sports events 246; sets of effects Japanese Olympic Games 59
244 Jeanrenaud, C. 247
Insights, Facebook: components of Jiangsu Hongshan Sports Park 223
engagement 151; content description Jiangsu Wutaishan Sports Center 223
151; data records 151; date 151; day of Jöreskog, K. G. 39
the week 151; engagement 151; hashtags Journal of Sport Management 74
151; negative user 151; post characters Jung, H. 112
151; post type 151; time posted 151; total
reach 151 Kahle, L. 153
Instagram 15, 16, 90, 130, 163; pictures Kaiser–Meyer–Olkin (KMO) 190, 278
58 Kaldor–Hicks welfare criterion 248
institutions 51 Kang, J. 207
intentions 96; affiliation 100, 102; age Kannan, P. K. 70
group 100, 101; education 100, 106; Kannan, V. R. 221
income 100, 105; party attending 100, Kaplanidou, K. 275
103; race 100, 104; to use by gender Kaplan, R. 250, 251
100 kappa matrix 41
interactivity 56, 163–165, 167, 168; Katz, M. 6
dimensions 59; relationships 114 Keller, K. L. 258, 259, 261, 264
interestingness 115 Kenny, D. A. 34, 36, 37
interfactor correlations 44 key stakeholder 222
 Index 297

Kim, K. A. 5, 8, 13 Li, J. 221

Kim, M. 284 likability 128
Klein, A. 39 Likert-type scale 98, 206, 207, 227, 277
Kline, R. B. 25, 26, 30 Lima de Carvalho, I. 135, 136
Knoll, J. 127 Lim, X. J. 138
Kono, S. 13, 14 linear regression 11
Korea Golf Course Business Association LISREL 8.80 program 39, 277, 278
110 Liu, H.-W. 178
Korea is a football league (K League) 109 live interviews 112
Korea Ladies Professional Golf Liverpool FC 60
Association (KLPGA) 110–113, 120; local area network (LAN) 52
online stakeholders 116; purpose of Local Organising Committee (LOC) 245
study 113; social media usage 112–113; local spectators 247
total ties 115 local sport industry incubator 236
Korean Baseball Organization (KBO) 109 location-based social media information
Korean female golfer 112 17
Korean Golf Association (KGA) 110 Lock, D. 6
Korean Internet and Security Agency “lock-in” participants 193
(KISA) 111 Lomax, R. G. 29
Korean LPGA golfer 112 long-distance road running event
Korean National Golf Tour 111 212
Korean Professional Soccer League 10 longitudinal study 5
Kotler, P. 153, 264 loyalty 146
Kressmann, F. 207 LPGA 112; Eleven Groups Users 116,
Kumar, L. 75, 79, 81, 85 117; media-related users 116; networks
Kunkel, T. 5 116, 117–118; online stakeholders
Ku, Ok-Hi 110, 111 116; primary stakeholders groups 116;
kurtosis values 44 purpose of study 113; social media usage
Kwak, D. H. 185, 207 112–113; total ties 115
Kwok, L. 154
machine learning technology 58
labeled items 263–264, 265 MacIntosh, E. 276
labeling 263 MacKinnon, D. P. 37
Ladies Professional Golf Association Magnusen, M. 29, 32
(LPGA) 109 major sports events: key indicators 254,
Lamigo Monkeys’ followers 164, 168–171, 254; problems of the I-O analysis
170 246–247; reasons to conduct 246
Lance, C. E. 39–41, 45 Malhotra, A. 165
Langmeyer, L. 130 mall-intercept approach 98
language bias 90 Mangold, W. G. 131
Larcker, D. F. 44 Manson, M. 155
latent factor means 41–42 manufacturer strategy 188
latent moderated SEM (LMS) 39 marathons 205
latent variables: circles 26; ellipse 26 market-based operation pattern 220
layout design 53 marketers 187
leadership skills 187 marketing: agencies 51; campaigns 94;
leagues 163 research 2; strategies 76, 131, 132
LeBreton, J. M. 36, 37 marketing implications: digital
Lee, D. 154 technologies 64–65; eSport affirmation
Lee, W. M. 112 64; fans 65; hybrids 64; interactivity and
level-based framework 201 virtuality 64; skill levels 64; sport con-
Lévy-Mangin, J. P. 136 tents 66; sports clubs 65; sports events
lights systems 60 features 64
298 Index

Marketing Intelligence & Planning measurement model 44

(journals) 74 media audience 51
marketing literature 144 media contents 55
marketing managers 81 media coverage of golf, South Korea:
marketing mix: consumer’s choice 55; con- broadcasting rights 111; economic
temporary sport venues 53; defined 53; growth 111; free Wi-Fi access 111; high-
digital communication 54; interactivity speed Internet 111; KLPGA 111, 112;
56; media devices 54, 55; place and dis- sport-related news 111
tribution 53; promotion 54; sports fan’s media devices 54, 55
satisfaction 54; stadium experience 53; media-dominated sports consumer 146
telecommunications 54–55; traditional media-related nodes 122
and innovative tools 54 media-related stakeholders 114, 116
marketing obstacles: discussion 192–196; media-related users 121, 122
growth rate 182; measures 189–190; mediation 38, 38; sport management
participants 188; perceived value 184; researchers 34
price fairness 183–184; price sensitiv- mediators 24, 34
ity 186–187; price tiers in sport 188; media types: images 164; text 164; videos
purchase intentions 186; racing industry 164
182; research design and procedures 189; Medina-Rodríguez, R. 276, 284
research questions 187; results 190–192; Meenaghan, T. 136
satisfaction 184–186; sporting events men golfers 114, 116
182; statistical analyses 190 merchandise 56, 57
market segmentation 107; affiliation meta-analyses 3, 127
97–98; age 96; consumer groups 96; meta-event 65
demographics 95; education 96–97; methodological aspects of the articles 76,
gender 96; income 97; party composi- 77–80
tion 98; race 97; sports consumption 96 methodological quality assessment 72
Marquez, A. A. 95 metric invariance 40
Mascot Media 94 Miami Dolphins 60
Mason, D. S. 218, 219 Michahelles, F. 177
Massachusetts Special Olympics 252 micro-celebrities: beauty 134; character-
Match-Up Hypothesis 128, 129, 137, 138 istics 138–139, 139; commercial value
Matheson, V. A. 243 135; endorsements 125, 127–128; fashion
Matthes, J. 127 126, 134; fitness 134; fitness industry
maturation 7 126, 135–136; food 134; perceived
maximum likelihood (ML): method 29; influence 135; product’s benefits 135;
standardized loadings 33 and purchase intention 136–138; social
maximum likelihood robust error (MLR) media 130–134; social media account
42 125; socio-cultural change 126; targeted
maximum ratio cut-off 30 consumers 126; traditional celebrities
McCarthy, E. J. 131 135
McCracken, G. 129 millennial sport fans 3, 4, 42
McKay, B. 182 misspecification 28; error 33
McKay, K. 182 mobile devices 60
mean 278, 279, 284, 285 mobile telecommunication 55
mean and covariance structure (MACS) model comparison: good model fit 34
41 model error 42
mean-centering 38 model estimation 29–32; multilevel struc-
mean differences 41 tural equation models 29; normality 29;
Meaning Transfer Model (MTM) theoretical justifications 29, 30
128–130, 129, 138 model fit indices 29–32; absolute, relative,
means 267, 268 and predictive 30; absolute vs. relative
measurement error 13; correction, SEM 25 (incremental) 30; chi-square test 30;
 Index 299

discrepancy 30; estimation method multivariate non-normality 39

independent vs. estimation method multivariate normality 29
dependent 30; normed vs. non- Murray, D. 186
normed 30; population vs. sample 30; musicians 127
sample-size dependent vs. sample-size mutual benefits 146
independent 30; simple vs. complex 30;
theoretical justifications 29, 30 Nanjing Olympics Sports Center 223
model-fitting software 24 Nanjing Youth Olympic Games 6
model identification: algebraic determi- Naraine, M. L. 75, 79, 85
nation 29; covariance matrix 28; data National Basketball Association (NBA)
collection 28; estimation of parameters 16–17
28; just-identified 28; over-identified 28; National Federation of State High School
under-identified 28, 29; variance- Associations (NFHS) 94
covariance matrix 28 National Football League (NFL) 9, 16, 57
model modification: fit index 33; measure- National Hockey League (NHL) games 61;
ment model level 33 Hartford Whalers 4
model specification: causal relationship National Provincial Championship (NPC)
27–28; equivalent models 26; model 144
parameters 26; “must-have” procedure National Sport Organizations (NSOs) 146
26; researchers 26; structural equation National Sports Education Council 277
model (SEM) 25; theory-based data NCAA Division I university sports
analytical method 26; variables 26 departments 76, 81
moderating effect 8, 38, 45, 96 needs: sport marketing analyses 1–2
moderation: testing boundary conditions negotiation-buffer model 14
38 nested models 31, 45
moderators 24 net output 245
modification indices 33 net present value (NPV) 247, 248
Moe, W. 188 network size 115
Moher, D. 72 news updates 163
Moosbrugger, H. 39 New York City Marathon 212
Morquecho-Sánchez, R. 276, 284 New Zealand Rugby’s (NZR) 144
Mothersbaugh, D. L. 97 Nielsen Media 57
motives 96 Nike: a global sportswear brand 127
Mplus 6.1. 32 Nippon (Japan) Professional Baseball
Mplus 7.0. 43 Organization (NPB) 262, 263
Mplus software 29 nodes’ attributes 114
Mullin, B. J. 260 node’s centrality 117
Mullin, N. 182 node size 114
multi-group SEM 45, 46 NodeXL Pro 118
multilevel structural equation models: data non-affiliated spectators 99
are multivariate normal 29; population non-athlete 202
distribution 29 non-centrality 31
multimedia 55, 59 non-economic impacts 247
multiplayer games 58 non-endurance event participants 204
multiple goodness of fit measures 13 non-event participants 211, 212
multiple regression 11 non-financial measures 251
multiple tickets 98 non-financial metrics 250
multiplier game 244 non-normative adjustment index (NNFI)
multi-purpose sports complexes 219 278
multivariate analysis of variance non-parametric: approach 14–15; method
(MANOVA) 10, 99, 190, 191–192, 37
195, 208; regression analysis 11; service nonpattern stadium 259
quality 11 non-peer-reviewed journals 72
300 Index

non-product-related attributes 166–167, Pan American Games 275

175, 178, 259 paperand-pencil survey 98
non-profit organizations 187 paper-based survey 76
non-profit promoter 189, 192 Parent, M. 276
non-residents 245 Pareto analysis 234–235
non-runner’s attitudes 203 Parganas, P. 166
non-sport-related corporations 66 Park, I. 110
non-sport-related organizations 114, 116 partial least square (PLS): independent
non-sport settings 163 variables 13; SEM 13
non-traditional events 206 partial mediation 37
normality: kurtosis test 29; Mardia’s nor- participant-focused obstacle 183
malized multivariate skewness and 29; participants 98–101, 154, 188, 204–205,
scatter plot and Q-Q plot examination 243; satisfaction 275; willingness 103
29; skewness and kurtosis measures 29 party composition 98
normed chi-square 30 passion 65
normed fit index (NFI) 267 path coefficients 44; table 33
Norton, D. 250, 251 perceived ease of use 94, 100–106
not-for-profit organizations 250 perceived service quality 5
Nunnally, J. C. 208, 284 perceived usefulness 94, 100–106
perceived value 184, 186
objectives 251 perceptions 7
observed sample covariance matrix 34 performance evaluation 235
occasional spectators 260 performance, Facebook: posts 156–157;
Oculus Rift 61 reach 155–156
Off-season brand interaction 147 performance measurement systems 251,
Oliver, R. L. 186 252
Olympic Games 66, 89, 246, 275 perishability 58
O’Mahony, S. 136 Perreault Jr, W. D. 131
omnibus test 40 personal selling 131
O’Neil 275, 284 Perugini, M. 43
one-sample t-test 9 Petrick, J. F. 186, 189
one-tailed t-test 9 Philips 221
one-way ANOVA 10 photo views 159
on-field spectacle 51 picture-based content analysis 17
online 15; communities 65; stakeholders pilot studies 3; clearest results, measures
115, 116; survey 76 3–4; ideas and insights, researchers 3;
operation directors 227 preliminary testing 3; unanticipated
optimal model 34 problems 3
organization-owned pages 147 Pizzo, A. D. 10
organization-to-employee 145 pleasure 205
organizers 248 Ployhart, R. E. 5, 39, 41, 45
Osokin, N. 165, 178 PLS-SEM 14–15
Oswald, F. L. 39, 41, 45 polychoric correlations 277
outdoor-leaning sport facilities 237 Popp, N. 75, 79, 86, 89
outliers 277 popularity 128
The Outline of the Healthy China 2030 Plan post-engagement 157
218 post hoc Tukey tests 10, 99, 100
Ozdemir, M. S. 231 posting types 163
post-reach 157
Pacific League teams 262, 263, 267 P1-P4 research sample 149, 153–159
paired t-test 9 pre-and post-behavioral intentions 6, 8
pairwise comparison matrices 232–233 pre-game music and video 60
Pak, Se-Ri 109–110 price fairness 183–184
 Index 301

price-sensitive consumer 186–187 rapid-response marketing 58

price tiers in sport 188 Ratkai, M. 145, 157
principal component analysis 12 Ratten, V. 144, 156
problems identification 2 real-time communication 57
procedural justice 183 rectangles, variables 26
product advertisement 129 refinement 224, 227
product-event 59 regression analysis 11, 25, 39
production efficiency 248 regression-based approach 39
product-related attributes 166–167, 177, regular-season posts 173, 175
259 relationship marketing 146, 152, 163
professional sport teams 163 reliability analysis 33, 227, 228–229
profitability 223 reliability estimates 270, 270
profitable marketing strategy 135 repurchase intentions 186
program history 81 reranking 234–235
progressive model 200 research design: and procedures 189; and
promoters 182 tools 2
promotions 54, 163, 195, 271 research questions/hypotheses 2
Pronschinske, M. 144, 156 resident consumer benefits 248
proportional engagement 159 residents’ lifestyle: disruption of 247
proportional performance, Facebook: residual correlation matrix 30
engagement 158; posts 158–160; reach retailer strategy 188
157–158 return on investment (ROI) 218, 219
proportional reach 158–159 Retweets 115
proposed model 45 Rio Summer Olympic Games, 2016 61, 110
proposed research model 42, 43 Rocha, C. M. 284
psi matrix 41 Rolex World Ranking 110
4 P’s Marketing Mix framework 131 root mean square error of approximation
psychological continuum model (PCM) (RMSEA) 31, 269
200, 204, 208 Ross, S. 207
psychometric guarantees 286 rotated factor structure 278, 280–283
psychometric properties 44 R-squares: model 38; values 34
public relations 131 Rugby Football Union (RFU) 243
public’s engagement level 113 Rugby League 145
public spending 245 Rugged Maniac 182
public stadium management 218 running magazines 213
public tax support 246
public users 114 Saaty, T. L. 224, 231, 233
Pugh, G. 94 sales promotion 131
purchase decision process 203 sample-covariance matrix 30
purchase intentions 186 Samsung Gear VR 61
Python codes 113, 114 Sandy, R. 244
Python programming language 113, 115, Santos, T. O. 165, 167
116 Sasakawa Sports Foundation (SSF) 263,
267
qualitative methodology 76 satisfaction 5; customer 185; customer sat-
Quarterman, J. 11 isfaction & repurchase 186; experiential
quasi-experimental design 7, 8 goods/services and customer 185–186;
Questionnaire of Athletes Satisfaction in transaction-specific 185
the Organization of a Sport Event 276 Sato, M. 6, 15
QUNISPORT V.MX instrument 276 Satorra–Bentler (S-Bx2) scaling method
29, 39
race 97 scalar invariance 40, 41
Random Consistency Index (RI) 233 Scandinavian Health Spas 130
302 Index

Scarborough Research identified Digital single activities 252

Savvy consumers 96 single arrowhead 26
Scheffe’s method 10 Sirgy, M. J. 207
Schoormans, J. P. 207 Sivakumar, K. 188
Schumacker, R. E. 29 six-phase thematic analysis 151
Scopus 72 size 188
scores 207 Sky boxes 60
search engines 71 Sky Sports 62
seating comfort 53 slippery slope 40
secondary data analysis 3 smartphones 60
secondary education 98 Smith, A. 75, 77, 81, 83
secured ownership 51 Soccer 145
self-concept 202 soccer balls 52
self-congruity 200–203, 212 social atmosphere 201
self-containment 27 Social Responsibility Journal 74
self-determination theory (SDT) 182 Social Identity Theory (research) 74
self-image congruity 207–209, 211 socialization opportunities 202
self-values 278, 280–283 social marketing 144
selling 65 social media 52, 60, 65, 66; analytics
sense of achievement 202 16–17; applications 60; with brand
sensory stimuli 165 attributes 164; communication 71;
Seoul Broadcasting System Golf (SBS competitive sports practice 88; con-
Golf) 110 sumption 88; decision-making process
Se-Ri Kids 110 125; defined 71; digital marketing 71;
service attributes 259 favorite athletes, teams, and events 112;
service capabilities 219 gender 88; golf courses 112; golf industry
service-led industry 145 in South Korea 109–111; KLPGA
service providers 65 and LPGA 112–113; management 70;
service provider selection: analytic marketing and communications 112;
hierarchical process (AHP) 224–227; marketing campaign 131; market-
discussion 235–237; purpose 222–223; ing strategies 125; method 113–114;
questionnaire 223–224; sports-focused micro-celebrities 130–134; online
219; stakeholder theory 222; subcontrac- environment 125; outlets 60; photos
tors 219; supplier selection and assess- and videos 112; purchase 88; purpose of
ment 220–221; supply chain 219–220 study 113; sport-related literature 148;
services-processes 252 stakeholders 112
SERVQUAL 275 social networking 60, 65, 88
seven-day week 151 social networks: analysis 114, 122; consum-
Shah, R. 132 ers 76; size 115; visualization software
Shanghai East Asia Sports Center 223 114
Shanghai Hongkou Soccer Stadium 223 social-psychological connection 201
Shanghai Xiang Cube Multi-Purpose social reality 113
Sports Complex 223 social spectators 58
shareholders 252 socio-demographic characteristics 89
Shaw, R. 244 software-based technologies 130
shoes 52 Song, D. 224
Shonk, D. 275, 284 soundboard 52
short-term marketing activity 146 Source Attractiveness Model 128, 129, 138
Siemens 221 Source Credibility Model 128, 129, 138
sign 205 South Korea: golf industry 109–111; inter-
signaling evidence 212 national sporting events 109; media
Simchi-Levi, D. 220 coverage of golf 111–112; social media
similarity 128, 138 usage 112–113
 Index 303

Spartan Race 182 types 163, 169, 170; professional sport

Spartan World Championship 182 organizations 163; social media users’
specification errors 36, 37 attention 164
specification model: causal relationship sport fan model of goal-directed behavior
27–28; equivalent models 26; model (SFMGB) 26
parameters 26; “must-have” procedure sport fans’ internet search 15
26; researchers 26; structural equation sport game attendance behavior 42
model (SEM) 25; theory-based data sport industry 1–2
analytical method 26; variables 26 sporting: equipment suppliers 51, 52; tech-
spectator dysfunctional behavior (SDB) nical partners 52
8–9 sport involvement process 200
spectators 52; affiliations 97–98; sport management: global reference jour-
convenience fees 100–106; data 17; nal 74; journals 13; researchers 29, 30;
demographic attributes 94; demo- scholars 31, 34; support model
graphic profiles 98; digital ticketing specification 32
94, 95; dissatisfaction 94; energy and sport marketers: characteristics 202;
happiness 8; motivations 10; with consumer information 2; and organiza-
multimedia and interactive dimensions tions 11; and problem-solving skills 1;
59; online purchases 93; perceived ease promotion campaigns 88; questions 2
of use 100–106; perceived usefulness sport marketing: business 1; communica-
100–106; quality 276; reporting 97; tion, effective 2; consumer variables 12;
trust of digital ticketing 100–106; data, analyze and interpret 2; digital
willingness 95 technology 15–18; needs 1–2; in North
sponsors 51, 65, 114, 116; golf industry in America 1; problems identification 2;
South Korea 111; image 52; partnerships research design and tools 2; research
114 questions/hypotheses 2; search and
sponsorship packages 94 comprehend related literature 2; social
sport activities: involvement theory 200; science disciplines 8; structural equation
program 276 modeling (SEM) 24–47; target popula-
sport businesses 67 tion 2; traditional analytical procedures
sport clubs 51; marketing mix 53 9–15; t-test 9
sport complexes 227 Sport Marketing Quarterly (article) 74
sport consumer 203; variables 12 sport nutrition 126
SPORTDiscus (EBSCO) 72 sport-related companies 66
sport event marketing 53, 54; equipment sport-related internet analysis 17
suppliers 52; eSport 52, 53; fans and sports: consumption 96; content manage-
spectators 58; first satellite transmission ment 67; disciplines 276; organizations
59; implementation 51; marketing mix 122; physicals venues 65; players 114;
53–56; multiple actors 51; network of service management 276; in in South
collaborative relationships 51; on-field Korea 109; tourism-oriented 236
spectacle 51 sport-service-leading complex 236
sport events: athletes’ satisfaction sportswear apparel 126
275–286; with children 105; social sport team context: brand associa-
media communication 89 tions 260–261; favorability 261;
sport fandom 201 non-product-related attributes 260–262;
sport fan Facebook pages: brand attributes product levels 260; product-related and
165–166, 171, 171–172; categories and non-product-related 260–262; profes-
variables 168, 169; chi-square test 170, sional sport teams 261; uniqueness 261
170; data analyses 168; data collection SPSS v.24 277
168; descriptive statistics 169, 169; squares, variables 26
hierarchical regression analysis 174, 174; staff member 99
limitations 179; operationalization 167, stakeholders: enterprise operation 222;
168; post characteristics 164–165; post network of collaborative relationships
304 Index

51; online 115, 116; sport marketing 4; Super Rugby franchises 149
sport organizations 163; theory 222; see supplementation products 126
also social media supplier selection: and assessment 220–221
stand-alone application 63 supply chain 219–220; cooperation 224
standard deviations 267, 268 supply chain management (SCM)
standardization levels 219 219–220; large- and middle-sized sport
Standardized Path Coefficients 35, 36 facilities 222; logistics network 220;
standardized residuals 33 manufacture 220
standardized root mean square residual surpluses to labor 248
(SRMR) 30–31 surpluses to state business 248
standardized sports marketing information survey takers 277
studies 3 symbolic congruity 211, 212
star-making system 134
star player 166, 260 Tafesse, W. 165
Star Sports 62 Tanaka, J. S. 30
state-centered economic mechanism 223 Tan, K. C. 221
State Champs 94 target audience 71
Statistical Package for the Social Sciences targeted marketing materials 103
(SPSS 16.0) 208 target population 2
Stavros, C. 145 tax revenue 245
Stempel, G. 151 team association model (TAM) 258; five
stereotypical image 202 product-related attributes 260; head
stereotypical user congruity 205, 207 coach 260; star players 260
store type 188 team attributes scale 263–264, 266
Stout, L. A. 222 team brand association scale (TBAS) 258,
Strategic management:a stakeholder 260; attributes 260; classified 260; 11
approach (book) 222 labeled groups 260
strategic sensitivity 66 team identification 5, 42, 45
strength 262, 271 team performance 81, 260
structural equations model (SEM) 6, teams 163, 260
11–15, 276, 278; advanced SEM pro- technological ecosystem 58
cedures 34–42; covariance-based SEM technology-enabled process 70
14; definition 25; measurement error Tejedor, S. 75, 80, 81, 86
correction 25; measurement models 25; telecommunications 54, 55
model comparison 34; model estimation telephone questionnaire 76
and model fit indices 29–32; model television (TV) 51, 55
identification 28–29; model modifica- theoretical model 32
tion 33; model specification 25–28; PLS- Theoretical Model for Athletic Event
SEM 14; regression analysis 25; research Digital Ticketing Use 94, 95
problems 24; selecting measures to theoretical reasoning 33
report 32–33; sport management 24; theory-based data analytical method 26
theory-driven data analytical technique Thompson, A. 144, 145
25; type I error reduction 25; user- Thompson, M. 136, 138
friendly statistical software packages 24 Ticket Roar 94
structural model 44, 45; of the satisfaction tickets: holders 81; multiple 98; purchase
284, 286 and sharing 98; sales 100
subscribers 52 ticket sales 56; promotion 81
substantial marketing strategy 104 Ticket Spicket 94
success 260 time-proven application 236
Suh, J. C. 185 time-switchers 247
Sumitomo Mitsui 221 Tokyo Olympics, 1964 59
Summer Olympics, 1988 109 top-class sport 66
 Index 305

Tough Mudder 182 urban, comprehensive service support 236

trade and industry associations 3 usermentions 113–115
traditional analytical procedures: mul- users’ attention 163
tivariate statistics 10–13; partial least U.S. Men’s National Soccer team 8
square (PLS) 13–15
traditional endurance events 206 validity 33
traditional sport marketing analytics: valuable services within stadiums and
descriptive (survey) design 4–6; exper- arena 51
imental design 6–9; exploratory design value adds 144
3–4 value-oriented 248
Trail, G. T. 39, 95, 97, 186, 284 Valve’s Counter-Strike: Global Offensive
training programs 213 Pro League 63
training software 52 vandalism 247
transaction-specific satisfaction 185 Vandenberg, R. J. 5, 39–41, 45
transport 245 variables 26; academic 74, 75;
triangulation 148, 149 bi-directionality 27; categories and
Trifecta World Championship 182 168, 169; endogenous 26; exogenous
Trofimenko, O. 137 26; explanatory 169, 169; omission/
trophy status 51 inclusion 28; SFMGB 42; sport game
true experimental research design 7 attendance behavior 42; subjective
trust of digital ticketing 94, 100–106 norm 42; team identification 42;
trustworthiness 138 uni-directionality 27
Tsitskari, E. 75, 80, 87 variance: differences 31; Pareto analysis
Tsuji, Y. 186, 189 of 234
t-tests: one-sample 9; paired 9; two-sample variance-covariance matrix 28
9 variance inflation factor (VIF) 173
Tucker–Lewis Index (TLI) 31 Varimax rotation 12
Tukey’s HSD test 10 Venkatesh, V. 96
Turner, B. A. 75, 77, 81, 82 video game market 65
Turner, J. 132 video games 52
turnover multiplier 244 video views 159
Tweets 115 Villarejo-Ramos, Á. F. 18
Twitter 15, 16, 18, 76, 88, 112, 113, 115, Vinerean, S. 131
118, 130, 163 virtual communities 65
Twitter Communication Networks 119, 121 virtual reality (VR) 56; sport sector 61–63
two-sample t-test 9 visitors 243
two-way ANOVA 10 vividness 163–165, 167, 168
two-way communication channel 132
two-way transfer 130 Walker, M. 130
type I errors 38, 40; reduction 25 Wall, S. 150
type II errors 30 Walt Disney World Marathon field 213
typical deviation 278, 279, 284, 285 Wang, Y. 220, 221
typologies of digital communication: Warrior Dash 182
intrinsic dimensions 59; live sports in Watanabe, N. 166
stadium 60; technological framework 61 wearable devices 56
web analytics 16
UCINET 117 Web of Science 72
UEFA European Championships 246 web-scraping 16
Ultra World Championship 182 website content guidelines 88
uni-directionality 27 websites 55
uniqueness models 262, 271 Weedon, G. 182
univariate analysis 208, 209 weighted least squares (WLS) 29
306 Index

Winter Olympics, 2018 109 Yim, B. H. 3, 4, 10, 12, 26, 39, 42

women golfers 114, 116 Yi, Y. 185
women’s sports organization 76 young fans 88
Word of mouth (WOM) 133 YouTube video clips 15
word-of-mouth referrals 184–185 Yu, B. 154
The Work Programme on Further
Reinvigorating Stadium Resources 218 Zanini, M. T. 75, 80, 87, 89
World Championships 243 Zeithaml, V. A. 184
World Cup 109, 243 Zeng, L. 221
World Sport Universiade 275 Zhang, J. J. 6, 11, 13–15, 29
Zhejiang Hangzhou Gymnasium 223
Xia, L. 189 Zhejiang Huanglong Sports Center 223
Xue, H. 218, 219 Zhejiang Shaoxing Olympic Sports Center
223
Yaakop, A. 153, 159 Zhejiang Taizhou Sports Center 223
Yazıcı, T. 4 Zhu, Y. Q. 132