Table 10. (Continued) Table 10. (Continued) Table 10

This article consists of five chapters. The Internet has revolutionized communication, allowing individuals and organizations to overcome geographical and time constraints, which in turn allows consumers and companies to connect around the world at any time. Online communities allow people to gather together on the Internet for various reasons, including searching for and sharing information, discussing communal issues, and making inquiries. In this article, you will be able to learn about Social media and digital marketing in the hospitality industry.

ii
Social media and
Digital marketing
in the hospitality
industry:

iii
Table of Contents
LIST OF FIGURES ..................................................................................................................
v
LIST OF TABLES ...................................................................................................................
vi
ACKNOWLEDGEMENTS ....................................................................................................
vii
ABSTACT .............................................................................................................................
viii
CHAPTER 1. INTRODUCTION ............................................................................................ 1
Background .......................................................................................................................
1
Problem Statement ............................................................................................................
6
Study Objectives ...............................................................................................................
9
Definitions of Terms .........................................................................................................
9
CHAPTER 2. REVIEW OF LITERATURE .........................................................................
11
Social Media and Online Community Marketing ...........................................................
11
Online Communities .......................................................................................................
13
Definition of an online community .........................................................................
13
Characteristics of an online community .................................................................
15
Theories to Explain Participation in an Online Community ..........................................
18
Economic theory .....................................................................................................
19
Social theories .........................................................................................................
20

iv
Online Community Participation ....................................................................................
21
Participation Benefits ......................................................................................................
25
Functional benefits ..................................................................................................
29
Social benefits .........................................................................................................
31
Psychological benefits ............................................................................................
32
Hedonic benefits .....................................................................................................
34
Monetary benefits ...................................................................................................
35
Outcomes of Online Community Participation ..............................................................
36
Brand commitment of online community members ...............................................
36
Brand Trust among Online Community Members .................................................
38
Moderating Role of Demographic Characteristics .........................................................
40
Moderating roles of age ..........................................................................................
41
Moderating roles of biological gender ....................................................................
42
Research Model ..............................................................................................................
44
CHAPTER 3: RESEARCH METHODOLOGY AND DESIGN ...........................................
47
Selection of Online Communities in Facebook ..............................................................
47
Sample ............................................................................................................................
48
Survey Instrument ...........................................................................................................
48
Data Collection ...............................................................................................................
53

v
Data Analysis ..................................................................................................................
54
Measurement model ................................................................................................
54
Structural model ......................................................................................................
55
CHAPTER 4: RESULTS ........................................................................................................
56
Demographic Characteristics ..........................................................................................
56
Brand Profile and Manipulation Check ..........................................................................
60
Descriptive Statistics for Measures.................................................................................
62
Measurement Model .......................................................................................................
66
Confirmatory factor analysis (CFA) for the hotel study .........................................
66
Confirmatory factor analysis (CFA) for the restaurant study .................................
72
Structural Model .............................................................................................................
78
Testing the structural model for the hotel study .....................................................
78
Testing the fully recursive model for the hotel study .............................................
81
Testing for moderating effects of age and biological gender for the hotel study ...
85
Testing the structural model for the restaurant study .............................................
90
Testing the fully recursive model for the restaurant study .....................................
93
Testing for mediating effects ..................................................................................
98
Testing the moderating effects of age and biological gender
for the restaurant study ...........................................................................................
99 Summary ...............................................................................................................
102

vi
CHAPTER 5. DISCUSSION AND CONCLUSIONS .........................................................
107
Discussion of Findings ................................................................................................. 107
Relationships between participation benefits and community participation ........
107 Relationships between community participation, brand trust, and brand
commitment ..........................................................................................................
110
Moderating effects of age and biological gender ..................................................
113 Additional paths between participation benefits, brand trust, and brand
commitment ..........................................................................................................
115
Conclusions and Implications ....................................................................................... 118
Managerial implications for the hotel study .........................................................
119
Managerial implications for the restaurant study .................................................
123
Summary ...............................................................................................................
126
Limitations and Future Study ....................................................................................... 128
APPENDIX A: A QUESTIONNAIRE FOR THE HOTEL SEMENT ................................
131
APPENDIX B: A QUESTIONNAIRE FOR THE RESTAURANT SEMENT ...................
135
APPENDIX C: APPROVAL OF THE USE OF HUMAN SUBJECTS ..............................
140
REFERENCES .....................................................................................................................
141
LIST OF FIGURES
Figure 1. Concepts of the virtual community ................................................................. 18
Figure 2. Proposed conceptual model for development of an effective online
community ....................................................................................................... 44
Figure 3. Moderating effects of age ................................................................................
45
Figure 4. Moderating effects of gender .......................................................................... 46

vii
LIST OF TABLES
Table 1. Definitions of online communities in the 21st Century ................................... 14
Table 2. Categories of community participants ............................................................. 23
Table 3. Reasons for participating in online communities ............................................ 27
Table 4. Community benefits from participation .......................................................... 29
Table 5. Constructs and items of the survey .................................................................. 52
Table 6. Demographic characteristics of the hotel sample ............................................ 58
Table 7. Demographic characteristics of the restaurant sample .................................... 59
Table 8. Brand profile of the sample ............................................................................. 61
Table 9. Perceived success of Facebook pages.............................................................. 61
Table 10. Descriptive statistics for all items used to measure model constructs ............. 63
Table 11. Correlation coefficients of constructs: initial measurement model
for the hotel study ............................................................................................ 68
Table 12. Correlation coefficients of constructs: final measurement model
for the hotel study ............................................................................................ 69
Table 13. Item measurement properties for the hotel study ............................................ 70
Table 14. Correlation coefficients of constructs: initial measurement model for the
restaurant study ................................................................................................
74
Table 15. Correlation coefficients of constructs: final measurement model for the
restaurant study ................................................................................................
75
Table 16. Item measurement properties for the restaurant study .....................................
76
Table 17. Summary of support for hypotheses based on the results of SEM in the
conceptual model (hotel study) ....................................................................... 81
Table 18. Chi-square test of model comparison for the hotel study ................................ 82
Table 19. Unstrandardized path coefficients and t-Values for structural model
(hotel study) ..................................................................................................... 83
Table 20. Moderating effects of age on the relationship between participation benefits
and participation in hotels‘ Facebook pages ...................................... 88
Table 21. Moderating effects of biological gender on the relationship between
participation benefits and participation in hotels‘ Facebook pages .................

viii
90
Table 22. Summary of support for hypotheses based on the results of SEM in the
conceptual model (restaurant study) ................................................................ 93
Table 23. Chi-square test of model comparison for the restaurant study ........................ 94
Table 24. Unstandardized path coefficients and t-Value for structural model
(restaurant study) ............................................................................................. 97
Table 25. Mediating effects of brand trust in restaurants‘ Facebook pages .................... 99
Table 26. Moderating effects of age on the relationship between participation benefits
and participation in restaurants‘ Facebook pages ............................ 100
Table 27. Moderating effects of biological gender on the relationship between
participation benefits and participation in restaurants‘ Facebook pages ....... 101
Table 28. Result of hypotheses tests for the hotel study ................................................
103
Table 29. Result of hypotheses tests for the restaurant study ........................................ 105

x
ABSTACT
Online community is an effective tool for building the relationship with consumers.
Many hospitality firms (i.e., hotels and restaurants) have utilized online communities a new
marketing channel to reach their consumers. Previous studies have identified four
participation benefits (functional, social, psychological, and hedonic) in the member
participation of community activities. In addition to these four factors, this study also added
monetary benefit as a predictor of member participation. Demographic factors (i.e., age and
biological gender) were proposed to influence the relationships between benefits and
community participation. As results of member participation in online communities, trust and
commitment toward hotel or restaurant brands have been considered as important factors that
enhance consumer relationships with hospitality brands. The purpose of this study was to
investigate benefit factors of member participation and the relationships between community
participation, brand trust, and brand commitment in hotel and restaurant online communities.
The present study investigated the conceptual model in two contexts, including hotel
and restaurant Facebook fan pages. After conducting confirmatory factor analysis, the
present study identified four benefit factors (functional, hedonic, monetary, and

socialpsychological benefits) as the predictors of member participation in hotel and
restaurant Facebook fan pages. Structural Equation Modeling (SEM) was used to test the
conceptual model.
xi
Based on the results of SEM, hotel and restaurant studies showed different results.
The results of the hotel study indicated that three benefit factors (functional, hedonic, and
social-psychological benefits) positively influenced members‘ community participation;
member participation positively influenced their trust toward a hotel brand. Biological
gender had a significant moderating effect on the relationship between functional benefits
and community participation in the hotel study. The results of the restaurant study indicated
that two benefit factors (hedonic and social-psychological benefits) positively influenced
members‘ community participation; member participation positively influenced their trust
and commitment toward a restaurant brand; members‘ brand trust also positively influenced
their commitment toward the restaurant brand.
The findings of this study provide significant insights for the researchers and
marketers. From the theoretical perspective, this is the first empirical research that
investigated consumer benefits and responses (i.e., community participation, brand trust, and
brand commitment) in online communities managed by hospitality firms. Thus, the study
contributes to the understanding of consumer behavior in social media. From the practical

perspective, the study suggests some strategies to effectively design hotel and restaurant
Facebook fan pages, which can strengthen the relationships with current consumers and
attract potential consumers.

1
CHAPTER 1. INTRODUCTION
Background
The Internet has revolutionized communication, allowing individuals and
organizations to overcome geographical and time constraints, which in turn allows consumers
and companies to connect around the world at any time (Harris & Rae, 2009). Online
communities allow people to gather together on the Internet for various reasons, including
searching for and sharing information, discussing communal issues, and making inquiries
(Wang & Fesenmaier, 2004a). With rapid changes in information technology, these online
activities are now performed via a new form of communication technology known as
‗Web 2.0‘ or social media (Gretzel, Kang, & Lee, 2008).
Social media are defined as ―a second generation of Web development and design,
that aims to facilitate communication, secures information sharing, interoperability, and
collaboration on the World Wide Web‖ (Paris, Lee, & Seery, 2010, p. 531). There are
numerous social media sites; among the most popular are Facebook, LinkedIn, and Twitter
(Jain, 2010). In 2010, Facebook announced it had over 500 million users; in the same year,
Twitter reported 75 million users (Paris et al., 2010; Owyang, 2010). On average, consumers
spend more than 5.5 hours per day participating on social media Websites (Nelsonwire,
2010). With their increasing use, these sites are perceived as tools for creating online
communities of users who share interests, activities, and objectives (Bolotaeva & Cata,
2010).
Many companies view the use of online communities as a profitable marketing tool
from which they can derive several benefits. First, companies can obtain vast amounts of
feedback regarding their products and brands by monitoring consumers‘ online
conversations, thus enabling them to resolve problems quickly and work to improve future
brand performance (Madupu, 2006). Second, online communities provide a real-time
resource regarding market trends and consumer needs. Companies can use these resources to
modify advertising messages and develop special targeted features for future products. Third,
companies can observe whether their brands are truly suited to consumers‘ lifestyles and can

2
learn which features of their products make them special or unique in consumers‘ eyes
(Kozinets, 1999). Through online communities, companies allow consumers to become
involved, directly or indirectly, in creating new products and brands (Sawhney & Prandelli,
2000). Overall, the popularity of online communities has heavily influenced many firms‘
marketing activities in recent years.
A brand community is comprised of consumers who are interested in a specific brand
(Jang, Ko, & Koh, 2007). There are two types of online brand communities: consumer-
initiated and company-initiated. As the names suggest, a consumer-initiated brand
community is developed voluntarily by consumers, whereas a company-initiated community
is sponsored and developed by a company. In a consumer-initiated community, consumers
benefit from uncensored feedback from other members (Jang, Olfman, Ko, Koh, & Kim,
2008). In contrast, a company sponsoring an online community may control the content
posted by its members. For instance, a message may be deleted if it contains negative
consumer opinions and experiences. If consumers recognize these actions, the company may
fail to build a strong online community because consumers can lose trust in the company and
its brand because of the perceived lack of transparency. A company should seek to foster
high levels of trust in and commitment to its brands in company-sponsored online
communities, factors that are more critical to the company than in consumer-initiated
communities (Jang et al., 2008).
Through participation in a company-sponsored online community, consumers can
compare products or services, share experiences with other users of the products, and suggest
alternative product choices. Moreover, consumers who participate in company-sponsored
online communities are often able to obtain exclusive information and special deals
(Antikainen, 2007) offered by the company. In such communities, companies can enhance
their relationships with consumers by providing special benefits that consumers consider
important (Antikainen, 2007). Through online member activities, positive attitudes about
other members of the community may be generated, and a sense of belonging can develop.
This further encourages consumers to share their experiences about the company‘s products,
especially when they have compliments or complaints (Madupu, 2006). Because of the
benefits of participating in online communities, a growing number of consumers join
company-sponsored online communities before making purchasing decisions (Muniz &
O‘Guinn, 2001).

3
Researchers have emphasized that community members‘ active participation is
critical in ensuring an online community‘s long-term survival (Madupu, 2006). Consumers
may be dissuaded from joining online communities if they do not see active communication
among the members and company. If the communities do not provide useful information
about products or brands, then consumers may show little interest in joining (Preece,
Nonnecke, & Andrews, 2004). Conversely, online communities with actively participating
members can attract new consumers and entice existing members to visit the community
more frequently or for longer periods (Preece, 2000).
In order to build and manage an active online community, companies first need to
understand their members‘ motivations with regard to the benefits that they expect in return
for their participation (Wasko & Faraj, 2000). If companies provide the desired benefits such
as information and a sense of belonging, they will be able to attract new consumers, build
relationships with them, and motivate them to visit again (Antikainen, 2007; Dholakia,
Bagozzi, & Pearo, 2004).
By building an active and effective online community, companies can foster strong
trust in and commitment to their brands. The majority of information and content in a
consumer-based online community results from consumers‘ experiences with products,
particularly with regard to their quality, maintenance, and directions for use (Muniz &
O‘Guinn, 2001). When the members collect information about a product from an online
community, they then anticipate that the products will perform as expected based on the
information provided by other members. When the members continuously experience
positive product performance and perceive the information to be trustworthy, they are more
likely to develop trust in the brand. Trust develops from shared beliefs about information
provided by community members and expectations of reciprocal communication (Blau,
1964).
In addition to trust, online community members can build commitment through
continuous participation. McWilliam (2000) revealed the impact of online communities on
building strong relationships between companies and their consumers. These strong
relationships reflect members‘ psychological attachment to the community and mutual belief
in each other (Morgan & Hunt, 1994). Commitment, like trust, can be enhanced as members
increasingly rely on the Internet for product information that will help them make purchasing
decisions (Shankar, Smith, & Rangaswamy, 2003). Reciprocal communication regarding

4
consumption experiences with brands enhances consumers‘ brand involvement and brand
commitment, especially when the products perform as expected based on the information
obtained (McAlexander, Schouten, & Koenig, 2002). Companies can therefore utilize online
communities to establish a strong bond with active community participants, which in turn can
generate trust and commitment to their brands (Ulusu, 2010).
While online community-building is a relatively new marketing strategy, its usage has
increased dramatically (Sweeney, 2000). The emergence of online communities has
stimulated researchers‘ and practitioners‘ interest regarding ways to accommodate these types
of communities into new business models. However, few researchers to date have empirically
investigated whether member benefits influence the level of participation in these
communities (Muniz & O‘Guinn, 2001; Wang & Fesenmaier, 2004a) as well as positive
consumer behaviors such as loyalty and contributions to the community (Kim, Lee, &
Hiemstra, 2004; Wang & Fesenmaier, 2004b). Their findings indicate that members spread
useful information about products and brands to other members and/or to their friends and
thus increase community participation (Kozinets, 2002). The relevance of online
communities for marketers is that active participation may create high levels of trust among
members as well as loyalty to the brand (Koh & Kim, 2004).
The majority of marketers would agree that the operation of a successful online
community is now highly relevant to successful marketing activities for many companies, yet
studies regarding online communities have been rarely conducted in the hospitality industry.
For this reason, the present study proposes the necessity of identifying and understanding the
factors that attract consumers to online communities for hospitality companies, and
investigating the relationships between consumer participation, consumer trust in brands, and
commitment to brands. From a theoretical perspective, this research provides an enhanced
understanding of consumers‘ motives for online social exchanges (e.g., Internet-based social
gatherings with other consumers and with a company) and of their cognitive processes during
the development of commitment to a particular brand. From a practical perspective, online
marketers can establish sustainable marketing strategies to keep online communities active,
identify what benefits community members look for in participating, and retain members who
are willing to be involved in a long-term relationship with the community.

5
Problem Statement
The present study focuses on how hospitality companies develop online communities
and which online platforms that they employ for building their communities. Despite the
proliferation of online communities in the hospitality industry, it is rare to find one that calls
itself an ―online hotel/restaurant community.‖ In fact, online communities launched by
hotels are commonly referred to as ―online travel communities.‖ For example, the Marriott
Corporation has launched an online travel community to replace its rewards program
(www.marriottrewardsinsiders.marriott.com). While a large number of hotels and restaurants
such as Hyatt and Marriott use social media as a platform for their online communities, they
are referred to as ―online travel communities‖ rather than ―online hotel/restaurant
communities‖ (Chkhikvadze, 2010).
Through social media, consumers share experiences with and suggest ideas to others
while developing new relationships within their communities. For this reason, many
hospitality firms consider social media a powerful tool to enhance consumer loyalty and
satisfaction (Kasavana, 2008). The results of a survey conducted by the Center for
Hospitality Research at Cornell University‘s School of Hotel Administration indicated that
approximately 25% of business travelers and over 30% of leisure travelers use social media
sites to read hotel reviews and obtain travel information before they make their travel plans
(Social Media, n.d.). Paris et al. (2010) indicated that Facebook is an excellent example of a
successful online community, with more than 500 million registered users around the world.
Given the number of users, upscale or boutique hotel properties in major tourism destinations
should create business Facebook pages to retain repeat guests and communicate with future
guests (Social Media, n.d.). Due to its worldwide popularity, Facebook was chosen as the
context of the present study.
A number of studies regarding online travel communities have identified the benefits
of member participation in online communities (Chung & Buhalis, 2008; Hwang & Cho,
2005; Wang, Yu, & Fesenmaier, 2002). Wang et al. (2002) identified four categories of
benefits: functional, social, hedonic, and psychological, and found that these benefits bear an
impact on whether members participate actively or passively. Although previous studies have
applied benefit constructs similar to those developed by Wang et al. (2002), the results of
these studies have been inconsistent, with diverging categories of benefits. These

6
discrepancies can occur due to the varied characteristics of online communities, such as
member characteristics, mutual member interests, and the communities‘ purposes (Kim et al.,
2004; Wang & Fesenmaier, 2004a).
The present study argues that, in addition to the four categories of benefits mentioned
above, monetary benefits influence member participation in an online community.
Consumers frequently seek monetary rewards from community participation (Seo, 2005).
Providing benefits of monetary value, such as discounts or coupons, may encourage the
participation of nonmembers, since economic value has been found to be a key element in the
initiation of a new relationship (Treadaway & Smith, 2010). Accordingly, the present study
employs previous benefit constructs specific to online communities and integrates the new
monetary benefit factor to investigate what members of a hospitality community seek to
obtain from their online interactions.
As mentioned earlier, the relationships between participation, trust, and commitment
to the community and to specific brands are important for the success of an online
community (Kim, Choi, Qualls, & Han, 2008; Ridings, Gefen, & Arinze, 2002). Studies have
found different outcomes from the relationships between these three components. For
example, Wu and Chang (2005) found a correlation between trust and member interaction,
indicating that each factor influences the other. Casaló, Flavián, and Guinalíu (2007) found
trust to be an outcome of member participation in an online community. Later, they showed
that trust is an antecedent of member participation (Casaló, Flavián, & Guinalíu, 2008) . Due
to the intangible nature of service, consumers in the hospitality industry tend to value
feedback based on other consumers‘ service experiences. Online community members are
more likely to search for information about hotels/restaurants before making a reservation for
rooms or tables and to compare their own service experiences to the information they
obtained from the community. If there are no discrepancies between their experiences and the
community information provided, members gain trust in the information obtained from their
community. The present study therefore proposes that trust is an outcome of member
participation (e.g., searching for information).
Study Objectives
The objectives of the present study are to (a) identify the benefits that participants in
online communities seek, (b) examine the relationships between members‘ levels of
participation, brand trust, and brand commitment, and (c) investigate the moderating effect of

7
demographic characteristics (i.e., gender and age) on the relationship between participation
benefits and community participation.
Definitions of Terms
Throughout the present study, the following terms are utilized for the purpose of
conceptualizing social media marketing and defining user behaviors:
Brand commitment: Strong and positive psychological attachment of consumers to a specific
brand (Beatty & Kahle, 1988).
Brand trust: Consumer confidence in a brand‘s reliability or ability to perform its stated
function (Ha & Perks, 2005).
Functional benefit: Value derived from achieving specific purposes (i.e., transactions,
information gathering and sharing, and convenience and efficiency) (Wang &
Fesenmaier, 2004a).
Hedonic benefit: Hedonic consumption experiences on the Internet that form creative
stimulation, positive emotions that are closely affiliated with feeling good, enjoyment,
excitement, happiness, and enthusiasm (Wang & Fesenmaier, 2004a, p.712).
Monetary benefit: Monetary advantages (i.e., monetary rewards such as discount coupons or
special deals) from relationships with a service provider (Gwinner, Gremler, &
Bitner, 1998).
Nonmonetary benefit: Time saved in searching for information (Gwinner, Gremler, & Bitner,
1998).
Online community: A group of people who share their consumer experiences via social
media, including communicating with other members or the company regarding their
concerns and opinions and providing critiques of offered services (Rheingold, 1993).
Psychological benefit: Value derived from a sense of belonging to the community and a sense
of affiliation with other members (Wang & Fesenmaier, 2004a).

8
Social benefit: Value derived from building relationships and performing interactions such as
providing information to help-seekers and receiving help (Wang & Fesenmaier,
2004a).
Social media: ―Web-based services that allow people to create a public profile, share
the connection with other users, and view and traverse their list of connections
in common network‖ (Ulusu, 2010, p. 2949).
CHAPTER 2. REVIEW OF LITERATURE
This chapter reviews the recent literature on social media and online community
marketing, and provides the underlying theoretical foundations of characteristics of online
communities. The participation benefits of online communities, member participation, and
consumer trust and commitment to a specific brand are discussed. Online communities in the
hospitality industry are conceptualized, taking into account the current usage of social media
for creating companies‘ online communities. Studies of brand trust and brand commitment
are examined to elucidate why members choose to maintain or enhance their relationships
with a specific brand on which an online community is based.
Social Media and Online Community Marketing
Social media are innovative Web-based applications in online marketing (Yang, Kim,
& Dhalwani, 2008). Companies utilize social media to form online communities to (1) build
new business models that include a new product marketing channel (Chung & Buhalis, 2008;
Ulusu, 2010; Yang et al., 2008), and (2) build strong relationships with consumers by
overcoming limitations of time and place (Bolotaeva & Cata, 2010; Sigala, 2003).
As a new marketing channel, online communities allow marketers to (a) gather
information about potential or current consumers from their profiles, (b) infer consumers‘
needs and preferences based on their history of community usage, and (c) obtain direct
replies from consumers (Sigala, 2003). Marketers can achieve a high level of customization
by monitoring content posted by community members and can obtain an in-depth
understanding of each consumer‘s needs, using this information to develop new
products/services. This helps marketers to advertise their new offerings to targeted consumers
(Chung & Buhalis, 2008).

9
Marketers view online communities as effective tools for building strong relationships
with consumers. These relationships can be enhanced further by the concept of ―Website
stickiness.‖ The ―stickiness‖ of a site is defined as its ability to draw and retain consumers by
creating consumer value, such as rewards for loyalty, personalized or customized
products/services, and trust (Zott, Amit, & Donlevey, 2000). Website stickiness can
encourage consumers to interact more often with other members of the online community and
with the company (Sigala, 2003).
Although social media provide companies with various marketing opportunities by
enabling them to build online communities, negative outcomes may arise with regard to
privacy concerns (Spangler, Hartzel, & Gal-Or, 2006). Social media encourage people to
provide personal information. In some cases, however, people may fail to take potential risks
into account, such as disclosing their information to the public. Details such as contact
information, age, and other specific information can be misused or can result in identity theft
by employees or third-party outsourced companies (Han & Maclaurin, 2002).
Despite privacy concerns, social media are nonetheless perceived as excellent
platforms for building a firm‘s online community because of the above-mentioned marketing
advantages (Sigala, 2003). In order to take advantage of online community use for marketing
purposes, a company must determine its target consumers and learn what motivates them to
visit its online community (Wang & Fesenmaier, 2004a). With the increasing usage and
popularity of online communities, most major companies no longer question whether they
should build online communities through social media.
Online Communities Definition of an online community
Although much research has been conducted regarding online communities, there is
still no generally accepted definition of the term ‗online community.‘ Among the various
definitions of online communities shown in Table 1, similarities drawn from these definitions
include that: (a) communication technologies (e.g., chat rooms, e-mail, and bulletin boards)
are the first prerequisite for the existence of online communities and (b) member
communication and interactions are functions of relationship building (Ä kkinen &
Tuunainen, 2005; Lee, Vogel, & Limayem, 2003). Considering these aspects, Lee (2005)
defined an online travel community as a collection of people who share interests in travel and
tourism, interact through online environments supported by advanced technologies, and

10
observe the shared values and norms of the online community. The present study uses the
following definition of online community for hospitality businesses: A group of people who
share their consumer experiences via social media, including communicating with other
members or the company regarding their concerns and opinions and providing critiques of
offered services.
Table 1. Definitions of online communities in the 21st Century
Researchers Definition
Jones & Rafaeli
(2000)
A symbolically delineated computer-mediated space where people
interact with each other by participating in and contributing to the
community.
Williams &
Cothrel (2000)
Groups of people who engage in many-to-many interactions online.
Preece (2001) A group of people who interact in a virtual environment. They have a
purpose, are supported by technology, and are guided by norms and
policies.
Balasubramanian
& Mahajan (2001) Any entity that exhibits all of the following characteristics: an
aggregation of people, rational utility-maximizers, interaction without
physical collocation.
Boetcher, Duggan, &
White (2002)
The gathering of people, in an online ―space,‖ where they
communicate, connect, and get to know each other better over time.
Ridings et al. (2002) Groups of people with common interests and practices, who
communicate regularly and for some duration in an organized way
over the Internet through a common location or mechanism.
Bagozzi & Dholakia
(2002) Mediated social spaces in the digital environment that allow groups
to form and be sustained primarily through an ongoing
communication process.
Lee et al., (2003)
A cyberspace supported by computer-based information technology
centered upon communication and interaction of participants to
generate member-driven content, resulting in a relationship being
built.

11
Kang, Lee, Lee, &
Choi (2007) A social group or organization, where people voluntarily become a
member and participate in interaction activities with other members
to exchange desired benefits they seek through a chosen community.
Note. Source: Lee (2005, p. 10)
Characteristics of an online community
Along with the various definitions of online communities, the characteristics of these
communities also vary across academic disciplines such as computer science, business, and
sociology (Wang et al., 2002). For instance, from a computer science perspective, Ellis,
Gibbs, and Rein (1991) characterized the online community as having two key components:
synchronous and asynchronous communication. Response time is the criterion that
distinguishes these components. Synchronous technologies, such as a chat room, require
people to be at their computers in order to communicate simultaneously, whereas with
asynchronous technologies such as bulletin boards and e-mail, people may respond to others‘
postings and take part in discussions at a later time. Online communities can provide both
synchronous and asynchronous technologies to support different communication tasks.
From a business perspective, Hagel and Armstrong (1997) identified three
components of an online community; a Webpage is published content, environment, and
commerce. Content published in an online community is the integration of members‘
communications based on specific topics. The Internet environment allows people to generate
and distribute their content without limitations of time and place. Companies can serve
commercial functions by facilitating online transactions in their online community.
Typaldos (2000) identified twelve elements of online communities drawn from
sociological theory: purpose, identity, reputation, governance, communication, groups,
environment, boundaries, trust, exchange, expression, and history. These twelve elements are
considered influential factors that lead to the success of a community. The first six elements
are based on individuals‘ needs and expectations of the community to which they belong; the
remaining six are related to the success of the community:
(1) Purpose: Members share a common goal and interest in the community.
(2) Identity: Members recognize other members‘ identities and create relationships.

12
(3) Reputation: Members build status based on their activities and others‘
expressions.
(4) Governance: The community controls members‘ behavior based on shared
values.
(5) Communication: Members interact with each other to share information.
(6) Groups: Members build small groups based on specific interests/tasks.
(7) Environment: Members interact in a synergistic environment, which enables
people to achieve their goals efficiently.
(8) Boundaries: Members are aware of those who belong to the community.
(9) Trust: Members trust other members and community organizers, leading to group
efficiency and problem-solving.
(10) Exchange: Members exchange resources, such as knowledge, goods, and
services.
(11) Expression: Members recognize how other members participate.
(12) History: Members keep track of past events and respond to them.
Wang et al. (2002) considered the sociological aspects of online communities, with
particular regard to the question of whether people apply the same social roles and
governance as those of physical communities. From theoretical and operational perspectives,
Wang et al. (2002)‘s specific functions and features of online travel communities are
described in Figure 1. The theoretical characteristics are place, symbols, and virtual. Place
involves more than communication technologies; rather, it is a physical community that
exists in members‘ minds. Symbols refer to the meanings and identities given to community
members. Virtual characteristics represent computer systems that influence how people form
communities. Wang et al. (2002) ‘s operational characteristics of an online community
include (a) people, who are the core of the community and actively perform activities; (b) the
purpose(s) shared by members and used to attract potential members, (c) the policies that

13
direct members‘ online behaviors; and (d) the computer system that makes this phenomenon
feasible in cyberspace.
In order to gain an understanding of what motivates individuals to participate in
online communities, the present study adopts Wang et al.‘s (2002) sociological perspective
regarding what encourages member participation in online communities. From this
perspective, the present study focuses on members‘ psychological mechanisms that determine
participation, and the influence of online community on members‘ social interaction (Bagozzi
& Dholakia, 2002). The present study also considers Hagel and
Armstrong‘s (1997) business perspectives for marketing practitioners, which explain that
people may extend their relationships with an online community for the purposes of finding
friendships, sharing common interests about particular products, gaining social support
regarding their consumption experience, and getting help in making purchasing decisions.
Based on the above discussion, the present study assumes that members decide to participate
based on the perceived benefits (i.e., to engage in activities that will help them achieve their
purposes such as gathering information, having fun, or making purchase).
Figure 1. Concepts of the virtual community
(Source: Wang et al., 2002, p. 410)
People Place Purpose
Virtual
Community
Symbol Virtual
Policy Computer
systems

14
Theories to Explain Participation in an Online Community
Various economic and social theories have explained why people visit online
communities: to gather information; to make transactions; to communicate and interact with
others; to have fun and experience enjoyment; to build new relationships; and to express
opinions and identity. All of these reasons for participation are contingent upon community
members‘ characteristics, shared purposes, and interests (Wang et al., 2002). In this section,
the reasons for individuals‘ participation are elaborated in light of theoretical explanations.
Economic theory
Online communities have gained attention from marketers and researchers due to their
economic power and their ability to affect power relationships between marketers and
consumers (Hagel & Armstrong, 1997). Because an online community is an aggregate of
consumers who show high interest in specific products or services, consumers who are
members of the community have greater intention to buy the products sold by the company
for which the community exists. Community members therefore can contribute to increased
profits for the company. These communities can also shift the balance of power from
company to consumers, because consumers are able to collect far more information than ever
before and their ideas influence the development and promotion of products (Butler, 2001).
A number of researchers have suggested that economic theory explains participation
in online communities (Gu & Jarvenpaa, 2003). Butler (2001) suggested the resource-based
model, which involves the concepts of perceived value defined by Zeithaml (1988):
consumer value will be created if consumers perceive more benefits gained than resources
sacrificed. The perceived benefits are the opportunities to obtain information resources and
knowledge from others, develop interpersonal relationships, etc.. The consumers sacrifice
time, attention, knowledge, and energy in order to receive these benefits. If the benefits
obtained exceed the sacrifices made, the community will provide value to its members, and
the number of participants will thereby increase (Butler, 2001). Similarly, Gu and Jarvenpaas

15
(2003) indicated that individuals will contribute only if they perceive more benefits than
costs, and that they are more likely to increase their participation when they recognize
incentives in the form of tangible or intangible returns. Member participation is significantly
related to the benefits that they expect to receive from the community. Therefore, providing
benefits that the members desire can encourage their participation.
Social theories
The present study employs two social theories (social exchange and social identity
theory) to elucidate members‘ motivations for social interaction within an online community.
Accroding to Blau (1964), social exchange is defined as reciprocal interaction among
individuals that benefits the involved parties. Individuals in these exchanges expect social
rewards (i.e., approval, status, and respect) through community participation. While there is
no guarantee for receiving anything for their contributions, individuals are willing to
contribute to the community as long as they can expect reciprocal interaction among
community members. That is, members A and B of a community (comprising members A–
Z) will provide help to members C and D without expecting gratitude or rewards from C or
D; however, they do expect to receive rewards from the community as a whole. Moreover,
the members who contribute to their community also expect to receive help from others when
they need it (Ridings et al., 2002). Social exchange theory explains that a higher level of
member interaction in the community will encourage the participation of others in
community activities (Blau, 1964).
Social identity theory explains how individuals identify themselves as members of a
group (Bagozzi & Dholakia, 2002). Social identity is a psychological state with cognitive,
affective, and evaluative components (Dholakia et al., 2004). The cognitive aspect of social
identity figures in the process of categorization, as individuals seek similarities with other
members and perceive dissimilarities with nonmembers. The affective component of social
identity involves emotional attachment or affective commitment to online communities
(Bagozzi & Dholakia, 2002). This emotional state influences the creation of loyalty and
citizenship behaviors (Meyer, Stanley, Herscovitch & Topolnytsky, 2002) and the retention
of relationships within the community (Bhattacharya & Sen, 2003). Finally, the evaluative
component is an individual‘s assessment of the value of being a member of the online

16
community (Dholakia et al., 2004). Members establish social identity based on the degree to
which they feel a sense of belonging to the community as well as the degree to which they
gain benefits from social interaction (Hogg & Abrams, 1988). When members identify
themselves as a part of the online community, they are likely to join and actively participate
in the community‘s activities (Dholakia et al., 2004).
Online Community Participation
Researchers have identified several categories of online community members, based
on levels of observation frequencies and community interactions (Table 2). Not all members
maintain the same level of interaction with other members and with the community as a
whole (Okleshen & Grossbart, 1998). ―Observation frequency‖ indicates the extent to which
members visit online communities but do not participate in community activities, whereas
―community interaction‖ denotes the extent to which members contribute to community
activities (e.g., sharing information and experiences; Lee, 2005).
Participation in online communities can be characterized as passive or active. Active
community members are those who interact with other members as opposed to those who
merely observe information (Madupu, 2006). Passive members browse online communities
but rarely become involved in community activities. Such members are referred to as
―lurkers or free riders‖ (Preece, Nonnecke, & Andrews, 2004). Because lurkers generate
traffic and increase Website hits, if online communities have a large number of members (i.e.,
both passive and active members), they tend to be successful. However, the numbers of
lurkers in an online community does not guarantee the community‘s success, given that these
members do not contribute to community activities. Rather, lurkers tend to pursue their own
goals and merely take advantage of the benefits of the communities (Ridings, Gefen, &
Arineze, 2006).
In contrast, active members are highly motivated to participate in online communities
and thus they are likely to share information and knowledge, contribute to fast dissemination
of valuable content to other members, and provide emotional support to other members
(Casaló et al., 2007). For instance, the popularity of YouTube is due to active members‘
considerable contributions to the community (Casaló et al., 2007). As community members

17
actively post product information and share experiences, the community acquires substantial
information that can attract new consumers and maintain strong relationships with existing
members. Furthermore, members‘ active participation enhances their knowledge regarding
brands and products (Muniz & O‘Guinn, 2001) and thus enables them to offer suggestions to
solve problems with product usage and help each other make purchasing decisions (Flavián
& Guinalíu, 2006). Active member participation is the key predictor of determining
community growth and ensuring the community‘s long-term survival (Koh & Kim, 2004).
Table 2. Categories of community participants
Authors Categories Description
Kozinets (1999), Wang &
Fesenmaier (2004a)
Tourist
Mingler
Has weak social ties with other members
Has somewhat strong social bonds with their
group and occasionally contributes to the
community
Devotee Strongly tied to the other members,
enthusiastic, and frequently participates in
community activities
Insider Maintains very strong bonds with other
members and very actively contributes to the
community
Burnett (2000), Preece et al.
(2004), Ridings et al. (2006)
Lurker ―Free-riders‖ who take advantages of the
community, but do not contribute to the
community
Poster
Participates in posting information and
messages and has higher willingness to
provide information and exchange social
support
Akkineu & Tuunainen
(2005)
Lead user Provides the necessary information to
develop new products for their community
Active user Provides valuable information for new
members

18
Researchers have found that community members‘ behaviors tend to evolve from
passive to active (Kozinets, 1999; Walther, 1996). New consumers may passively participate
in online activities to gather information and determine whether they share community
interests (Walther & Boyd, 2002). However, as consumers spend more time in a community
and the number of their interaction experiences increases, they are more likely to become
frequent users, perceive themselves as members, and eventually become active members of
the community and loyal consumers of the company‘s products (Preece et al., 2004). Thus,
understanding the evolution of online community member involvement helps marketers
segment their members into subgroups based on their level of participation (Preece et al.,
2004).
As discussed above, members‘ active participation in online communities is a key
element to ensure the growth and sustainability of these communities. In order to attract new
members and encourage existing members‘ active participation, online community marketers
must understand consumers‘ motivations to participate relative to what they desire to receive
from online communities. Understanding participation benefits is critical in order for online
community marketers to establish the optimal approaches not only to attract new members
but also to encourage non-active members‘ participation, which means converting lurkers
into active participants. Ultimately, companies that have online communities with a large
number of active members tend to become successful in building long-term relationships
with their consumers.
Participation Benefits
In order to build successful online communities, community marketers must attract
participants and encourage them to remain loyal to the community. One way to maintain
online community traffic is to provide members with specific benefits that they desire from
their participation as a community member (Wang et al., 2002). Kang et al. (2007)
emphasized that such benefits should be consistently provided. If the online community fails
to deliver consistent benefits to community members, the success of the online community
may be jeopardized (Wang et al., 2002). When members perceive the benefits as worthwhile,
they are more likely to become more active participants (Morgan & Hunt, 1994).
Table 3 shows that researchers have identified a variety of reasons that consumers
possess for participating in an online community, including motivational and benefit factors

19
(Armstrong & Hagel, 1996; Koh & Kim, 2004). The former reasons were developed by
Dholakia et al. (2004) while the latter were developed by Wang and Fesenmaier (2004a).
Dholakia et al.‘s (2004) study focused on the social influence of consumer participation in
online communities and found five motivational factors: purposive value, self-discovery,
entertainment, maintaining interpersonal interconnectivity, and social enhancement. The
findings explained communities‘ social influence on members through an understanding
participation benefits that they desire to obtain from social interaction. For example, in a
network-based online community where participants are not familiar with each other in most
cases, members seek to attain benefits based on their individual needs as related to purposive
value (e.g., obtaining information), self-discovery (e.g., expressing preferences), and
entertainment. In contrast, the social influence model suggests that a member‘s decision to
participate relies on other members‘ participation behaviors (i.e., intentional social action). In
other words, members may choose to participate only if they observe high levels of
interaction among other members (Dholakia et al., 2004).
In an online tourism community, members seek to accomplish a variety of travel-
related tasks such as gathering travel information, making transactions (e.g., booking travel
packages), anticipating new relationships with people in remote and international places, and
looking for individuals to accompany them on a backpacking or knapsack tour of Europe
(Hagel & Armstrong, 1997; Wang et al., 2002). According to Wang and Fesenmaier (2004a),
these can be considered as tourists‘ fundamental needs (i.e., human needs), and they have
been generally accepted and classified into four categories: functional, social, psychological,
and hedonic. Functional benefits are related to information gathering and transactional
processes; for example, online community members compare the quality of information
obtained with the amount of time and effort that has been invested in community activities.
Social benefits describe the development of relationships with other people through
communication and interaction. Psychological benefits refer to the emotional aspects of
relationships, such as a sense of belonging and affiliation with the community (Wang et al.,
2002). Hedonic benefits indicate a positive emotional state resulting from entertainment and
enjoyment (Wang & Fesenmaier, 2004a). These four benefits are discussed further in the next
section.
Table 3. Reasons for participating in online communities
Authors Benefits or needs

20
Hagel & Armstrong (1997) Transaction, interest, fantasy, relationship
Vogt & Fesenmaier (1998) Functional, hedonic, aesthetic, innovation, and sign needs
Wang & Fesenmaier (2004a) Functional, psychological, social, and hedonic needs
Kim et al. (2004) Membership, influence and relatedness, integration and
fulfillment of need, shared emotional connection
Hwang & Cho (2005) Functional, social, psychological needs
Chung & Bulahis (2008) Information acquisition, social-psychological needs,
hedonic needs
Wang and Fesenmaier (2004a) argued that the motivation for consumer participation
in an online tourism community relates to fundamental needs (i.e., participation benefits),
whereas Dholakia et al. (2004) contended that this motivation is determined by social
influence (i.e., the influence of other members‘ interaction on one‘s participation decision).
However, Madupu (2006) claimed that Dholakia et al.‘s (2004) motivation model can be
reconciled with Wang and Fesenmaier‘s (2004a). According to Dholakia et al. (2004),
motivational factors only take into account consumers‘ intentional social action in online
communities. That is, individual members tend to more actively engage in community
activities for purposive value (e.g., exchange information), self-discovery (e.g., expressing
preferences), interaction (e.g., making friendship), social support (i.e., emotional support),
and entertainment (e.g., recreation). Table 4 shows that Dholakia et al.‘s (2004) motives can
be related to benefit categories proposed by Wang and Fesenmaier‘s (2004a).
Wang and Fesenmaier‘s (2004a) framework is employed in the present study because
hospitality-related communities have features similar to travel communities. Members of a
hotel or restaurant community are likely to be involved with activities such as searching for
information about a property (e.g., the ambience of hotel or restaurant and the quality of
service), sharing service experiences with other members, and communicating with the
service provider. For example, a hotel guest may seek out other guests‘ experiences with a
property in the hopes of gaining more information about the neighborhood with regard to
sightseeing and restaurants (―Hotel News‖, 2008). Restaurant consumers can search for

21
information about menus and new promotions while making a decision to make reservations,
visit a restaurant, or place orders via a company‘s online community (Kasavana, 2008).
In addition to the four benefits identified by Wang and Fesenmaier (2004a), the
present study considers consumer desire for economic advantages from building a
relationship with a service provider (Harris, O‘Malley, & Patterson, 2003). The term
―monetary benefit‖ is adopted from Gwinner et al. (1998); this benefit can significantly
influence the extent of members‘ participation in online communities. Based on the
discussion above, the present study proposes that members hope to gain five specific types of
benefits from participation in the online community: functional, social, psychological,
hedonic, and monetary.
Table 4. Community benefits from participation
Category Benefit
Functional
Information/Purposive value*
Efficiency
Convenience
Psychological Affiliation
Belonging
Identification
Self-Discovery*
Social Communication
Relationship/maintaining interpersonal interconnectivity*
Involvement
Trust
Social enhancement*
Hedonic Entertainment*
Enjoyment
Amusement
Fun
Note. Source: Madupu (2006). * Motivational factors from Dholakia et al. (2004) related to
Wang and Fesenmaier‘s (2004a) benefits.

22
Functional benefits
A functional benefit is one that increases the ease and/or efficiency of completing
transactions (i.e., purchasing products and services) and exchanging information (i.e.,
information gathering and sharing) (Peter, Olson, & Grunert, 1999). One of the functional
benefits of an online community is that interaction with other community members can
facilitate purchasing decisions (Armstrong & Hagel, 1996). Information exchange is one of
the major reasons for online community participation (Arsal, Backman, & Baldwin, 2008).
Activities included in information gathering and sharing can be divided into two categories:
solving problems and sharing information with others (Nishimura, Waryszak, & King, 2006).

30
While searching for information, community members can obtain answers to their questions

24
or disseminate useful information to others, including families, friends, and other community
members (Wang et al., 2002). Convenience and efficiency can be realized through the
Internet since members can have easy access to a vast amount of information relevant to their
purposes with no temporal or geographic constraints. Since the information is stored and
accessible within online communities, members can search for and exchange information
more efficiently (Wang et al., 2002).
The relationship between functional benefits and community participation has been
well documented but inconsistent in tourism research. Wang and Fesenmaier (2004a) found
functional benefits to have a negative impact on members‘ participation. One reason for this
negative relationship may be that members enjoy the fun and interactive parts of the
community activities more than the task-oriented ones. Hwang and Cho‘s (2005) study
revealed no significant relationship between functional benefits and members‘ community
activities. In contrast, Chung and Buhalis (2008) found a positive relationship between
members‘ information acquisition and their participation. Although members might not have
specific plans for travel or dining out, they can still collect and share information about
destinations, hotels, and the best restaurants in the area. If members can achieve their specific
goals, such as acquiring information quickly, they are more likely to visit their online
community.
Based on the above discussion, the present study posits that the relationship between
functional benefits and community participation can vary depending on what community
users want to gain from the community (i.e., whether they focus on entertainment or
information acquisition). However, the present study postulates that members in specifically
hospitality-related communities have explicit needs to obtain information with regard to hotel
facilities, room rates, restaurants, and tourism information, especially when they are actively
planning a trip. They will also share their experiences with others, offering suggestions or
responding to questions. Thus, the following hypothesis, indicating a positive relationship to
stimuli members‘ active participation, is proposed:
H1: Functional benefits have a positive influence on online community participation.
Social benefits

25
Social benefits are the various kinds of help and support that members provide for
each other (Wang & Fesenmaier, 2004a). Community members help and support each other
by exchanging ideas and opinions of interest, answering other members‘ questions, and
introducing new topics for discussion (Dholakia, Blazevic, Wiertz, & Algsheimer, 2009;
Madupu, 2006). All of these activities can be enhanced when community members are highly
involved with each other. Frequent participation in community activities engenders trust
(Wang & Fesenmaier, 2004a), which in turn leads to flourishing relationships among
community members and with organizers of the community (Preece, 2000). For example,
sharing service experiences at a hotel or restaurant with other members and finding people
who have similar concerns and needs regarding room type or dinner menus require active
interaction among members who have similar interests and experiences. When individuals
recognize each other and identify the online community as their reference group, they are
more likely to contribute valuable information and support each other‘s activity (Preece,
2001).
The rapid growth of online communities around the world speaks to the popularity of
establishing and nurturing social interactions. As members spend more time online in the
community (Walther, 1996), this social interaction becomes a part of their lives (Feenberg &
Bakardjieva, 2004). Because the Internet enables people to overcome the limitations of time
and space on communication and interaction, individuals from different countries can join
together and contribute to the knowledge and information (Chung & Buhalis, 2008).
Previous studies have indicated that social benefits significantly influence members‘
attitude toward an online community (Chung & Buhalis, 2008; Wang & Fesenmaier, 2004a).
Coon (1998) found that the primary reason people choose to participate in online
communities is to build friendships with others who have similar interests or purposes.
Online community members tend to increase the number and length of visits to online
communities, and to actively participate in online community activities, when they recognize
that they share mutual interests with other members (Hwang & Cho, 2005). Based on these
findings, the following hypothesis is proposed:
H2: Social benefits have a positive influence on online community participation.

26
Psychological benefits
Psychological benefits are derived from feeling connected to community members,
and include identity expression through the community, a sense of belonging to the
community, and a sense of affiliation with other members (Bressler & Grantham, 2000).
According to Kozinets (1999), online community members can gain knowledge not only
about products or services but also about group norms, specialized language, and concepts
within the community (i.e., members‘ identities). As members gain such knowledge about
their online communities, they come to understand the community and feel a strong sense of
belongings and affiliation, which in turn develops a permanent sense of identification (Wang
& Fesenmaier, 2004a). Once members fully identify themselves as a member of the
community, they are more likely to rely on information provided by the community
(Anderson & Weits, 1989). This psychological dependence makes members feel confident
and positive about their interactions, a psychological benefit that encourages members to
increase their participation.
Bressler and Grantham (2000) indicated that psychological benefits are a starting
point for joining an online community due to an individual‘s need for a fulfilling sense of
belonging to a community. However, in one particular tourism study, no relationship was
found between psychological needs and member participation (Wang & Fesenmaier, 2004a).
Wang and Fesenmaier (2004a) provided an explanation for this result, stating that, in online
travel communities, members do not know each other well, and may not desire a sense of
community with other members or feel the necessity of developing member identification
(Dholakia et al., 2004). In another tourism study, Hwang and Cho (2005) indicated that
psychological benefits significantly influence members‘ attitudes toward the online
community, while Kim et al. (2004) found a positive relationship between sense of
community and members‘ loyalty to the community.
Although previous studies have obtained different results regarding this relationship,
more studies show that psychological benefits have a positive influence on online community
participation than not (Dholakia et al., 2004; Kim et al., 2004). Consumers may increase their
level of participation in order to express their preferences, which lead to the formation of an
emotional attachment with other members and the community (Lee, 2005). A sense of
community enables consumers to share experiences and solve problems related to

27
consumption (Bakos, 1998). This is an effective way to allure new consumers and retain them
as loyal consumers (Kim et al., 2004). Thus, the following hypothesis is proposed:
H3: Psychological benefits have a positive influence on online community
participation.
Hedonic benefits
Hedonic benefits include positive emotional states, such as feeling entertained and
amused and experiencing enjoyment that occurs when participating in community activities
(Wang & Fesenmaier, 2004a). In online communities, members are likely to engage in
activities that not only provide valued information but also elicit positive emotions (e.g.,
happiness, excitement, and enthusiasm) (Armstrong & Hagel, 1995). Some online
communities allow members to play games or participate in contests or polls related to
members‘ mutual interests, which lead to pleasure, fun, and entertainment (Wang &
Fesenmaier, 2004a). From a hedonic perspective, community members are viewed as
pleasure seekers, who place more value on the experiential aspects of consumption than on
other participation benefits discussed above (Vogt & Fesenmaier, 1998).
For some online community members, hedonic benefits are more important than other
benefits (Hoffman & Novak, 1996). Participation in an online community is influenced by
hedonic benefits that members gain from discussion forums, electronic bulletin boards, and
features for sharing pictures and videos (Dholakia et al., 2004). If participating in an online
community is perceived as fun or entertaining, members are more likely to visit the
community and to spend more time visiting it. Therefore, the following hypothesis regarding
hedonic benefits and community participation is proposed:
H4: Hedonic benefits have a positive influence on online community participation.
Monetary benefits
Consumers seek to receive economic advantages from a relationship with a service
provider, which can be referred to as monetary savings (Gwinner et al., 1998). Monetary
savings (i.e., discounts or special price breaks) is a primary reason for a consumer to develop
a relationship with a company (Harris et al., 2003; Peterson, 1995). Gwinner et al. (1998)
illustrated the importance of monetary benefits when developing a relationship with a service

28
company. In the following excerpt, they give an example of the monetary benefits gained by
loyal consumers:
Monetary benefit: I often get price breaks. The little bakery that I go to every
morning, every once in a while they‘ll just give me a free muffin and say,
―You‘re a good consumer, it‘s on us today.‖ Also, my hair stylist one year
said, ―Oh, it‘s your birthday, okay; I‘ll give you your haircut.‖ You‘re not
going to get that if they don‘t know you. (p. 104)
In hospitality research, monetary benefits have been considered a part of
individualized services that fulfill consumers‘ specific needs (i.e., special treatment benefits)
(Lee, Ahn, & Kim, 2008). Han and Kim (2009) found that special treatment benefits (e.g.,
gift certificates) had a positive effect on the way that consumers felt about a restaurant. A
similar process is likely to occur in online communities for hotels and restaurants (Kozinets,
1999). That is, offering monetary benefits is likely to have as positive an effect on online
community members as it does on actual patrons of a restaurant (Kozinets, 1999). These
businesses tend to offer special promotions and coupons to attract new members and benefit
online community members (Treadaway & Smith, 2010). For example, community events
and contests that provide winners with something of monetary value (e.g., coupons,
information about sales) may encourage member participation and entice nonmembers to
register with the community. Thus, monetary benefits attract new members and maintain
existing relationships. Therefore, the following hypothesis is proposed regarding the
relationship between economic benefits and community participation:
H5: Monetary benefits have a positive influence on online community participation.
Outcomes of Online Community Participation
In this section, the relationships between online community participation, brand trust,
and brand commitment are explored and hypotheses are developed. Through participation in
online communities, members provide helps to others and receive helps when they need it.
Because hospitality products and services cannot be evaluated without consumption,

29
consumers can be significantly influenced by others who have had experiences with those
products and services. Once consumers find information provided by other people to be
trustworthy, they learn to rely on these opinions (Paris et al., 2010). A high level of trust
fosters emotional attachments among members of online communities (Hagel & Armstrong,
1997; Hess & Story, 2005); it also increases their level of commitment to a particular brand
(Casaló et al., 2007).
Brand commitment of online community members
Consumers have been shown to engage different cognitive processes in evaluating
information about their preferred brands or competing brands (Raju, Unnava, &
Montgomery, 2009). The information selection process can be influenced by brand
commitment, which is defined as a strong and positive psychological attachment of
consumers to a specific brand (Beatty & Kahle, 1988). On the one hand, consumers who are
highly committed to a specific brand evaluate competing brands less positively or avoid
considering competitors‘ brands when making purchasing decision (Ahluwalia, Burnkrant, &
Unnava, 2000). They tend to defend their favorable attitudes toward brands when perceiving
a threat such as unfavorable information about their preferred brands or favorable information
about competing brands (Chaiken, Liberman, & Eagly, 1989). Consumers who perceive such
threats tend to secure their positive attitudinal position toward their preferred brands by
searching for favorable information about their brand (Jain & Maheswaran, 2000) and
maintaining their beliefs about the brands (Kunda, 1990). In other words, consumers want to
see evidence that their preferred brands are different from and better than other brands
(Chaiken et al., 1989).
On the other hand, consumers who are less committed to a specific brand are less
likely to be threatened by competing brands (Jain & Maheswaran, 2000). These consumers
are likely to consider any brand that satisfies their needs and to seek information about new
brands (Raju et al., 2009). They look for similarities between the positive aspects of their
preferred brand versus its competitors (Sanbonmatsu, Posavac, Vanous, & Ho, 2005). There
is a high possibility that these consumers may accept alternatives when they feel that the
competing brand is similar to their preferred brand (Kruglanski & Webster, 1996).

30
An online community often constitutes a group of committed consumers because the
group consists of people who share common interests and purposes (Bagozzi & Dholakia,
2002). Members are likely to discuss how to use products, and ask other members for product
repair and maintenance information (Casaló et al., 2007). As members frequently and actively
participate in online communities, they become more familiar with the brand, and thus
develop expertise on products and brands. These members also are likely to help other
members within the community (Muniz & O‘Guinn, 2001). Information or content posted by
these members contains positive messages in support of their favorite brands, which protects
their attitudinal positions about those brands (Raju et al., 2009).
Being highly involved in community activities (e.g., participating in discussions and
posting positive messages about a brand) positively affects commitment and emotional
attachment to a brand (Algesheimer, Dholakia, & Herrmann, 2005). Consumers‘ emotional
ties toward particular brands can develop as a result of active participation in online
communities (Casaló et al., 2007). For example, when consumers discuss common issues
related to their favorite brands, they are more likely to create emotional ties with each other,
and they reach agreement more easily. Active participation increases members‘ commitment
to particular brands because members who share similar interests in those brands can
communicate and interact with each other through community discussion boards. When they
experience shared sympathy on specific issues related to their preferred brands or
consumption experiences, positive attitude toward those brands can be enhanced
(Algesheimer et al., 2005; Wang & Fesenmaier, 2004a). Therefore, the level of participation
positively affects commitment to a brand. Thus, the following hypothesis is proposed:
H6: Online community participation has a positive influence on brand commitment.
Brand Trust among Online Community Members
Trust is a fundamental principle of interpersonal exchange, built up gradually through
repeated interactions (Gefen, 2000; Leimeister, Ebner, & Krcmar, 2005). Brand trust is
defined as consumers‘ secure belief that a brand will perform as expected upon consumption
(Ha & Perks, 2005; Pitta, Franzak, & Fowler, 2006). Trust is an essential element in reducing
perceptions of risk. When a brand successfully performs its expected function, consumers

31
begin to trust it and decide to continue a relationship with the company or brand (Butler &
Cantrell, 1994). Without trust, interactions may not continue beyond a single occasion
(Gefen, 2000).
According to Flavian and Guinaliu (2006), frequent participation in community
activities (e.g., posting and reviewing messages) enables consumers to be more
knowledgeable about brands. For example, consumers can discuss experiences of product
usage and suggest alternative ways to use or fix the brand‘s products (Flavian & Guinaliu,
2006). These discussions increase consumers‘ confidence that they will be satisfied with a
particular brand and thus build trust in that brand (Ha & Perks, 2005).
In addition, online communities serve as bulletin boards for posting consumers‘
opinions and suggestions, and companies consider these resources when developing new
products or modifying brand products (Casaló et al., 2007). Companies seek to utilize their
online community as a tool for exchanging ideas about new offerings, directly listening to
product/service comments from consumers, and learning more about consumers‘ needs.
Through continuous interaction between companies and consumers, their trust in the
company and its brands is eventually generated (Tung, Tan, Chia, Koh, & Yeo, 2001). Based
on this communication and interaction, consumers expect that the brand will constantly meet
or exceed their fundamental needs. When consumers are satisfied with what they receive
from participation, they may increase their levels of trust toward the online community and
the brand (Deighton, 1992).
In numerous marketing studies, trust has been identified as a major predictor of
consumers‘ long-term relationship with and commitment to a brand (Garbarino & Johnson,
1999; Harris & Goode, 2004; Morgan & Hunt, 1994). Brand commitment is referred to as
consumers‘ positive emotion toward a brand (Beatty & Kahle, 1988). Since committed
consumers are satisfied with the brand, they are less likely to look for other brands, which
will save them time and effort (Garbarino & Mark, 1999). Positive emotion toward a brand is
related to consumers‘ trust that the brand will perform its functions (Ha & Perks, 2005). In
addition, brand trust strengthens attachment and favorable behaviors toward brands (Beatty &
Kahle, 1988). Loyal consumers tend to avoid all other alternatives and rely on information
about their favorite brand (i.e., a tendency to resist changes) (Pritchard, Havitz, & Howard,
1999). Based on the above discussion, the following two hypotheses are proposed:

32
H7: Online community participation has a positive influence on brand trust.
H8: Brand trust has a positive influence on brand commitment of online community
members.
Moderating Role of Demographic Characteristics
Certain demographic characteristics affect the way online consumers behave (Morris
& Venkatesh, 2000; Serenko, Turel, & Yol, 2006). In particular, consumer age and biological
gender have been identified as influential determinants in an individual‘s behavior
(e.g., information searching, downloading and updating information, and
purchase/reservation transactions) (Matzler, Grabner-Krauter, & Bidmon, 2006; Saad & Gill,
2001). For example, younger consumers, between the ages of 20 and 30, use the Internet
frequently for chatting, emailing, meeting new friends, and playing games (Thayer & Ray,
2006), whereas older Internet users between the ages of 50 and 64 use it more often for
checking email and communicating with family members (Howard, Rainie, & Jones, 2001).
Previous studies have also found biological gender differences in Internet usage behaviors;
women are more involved with social relationships and prefer to maintain those relationships
more intimately than men (Boneva, Kraut, & Frohlich, 2006). Because age and biological
gender are associated with patterns of Internet usage behaviors, understanding the effects of
these two demographic characteristics on online communities is important (Igbaria &
Chakrabarti, 1990).
Moderating roles of age
Age has been considered as the most important personal characteristic that affects
computer adoption and Internet usage behaviors such as messaging, searching, downloading
information, and purchasing (Teo, 2001; Serenkoet al., 2006). Morris and Venkatesh (2000)
linked technology adoption with age differences. Younger individuals are more open to using
a new technology than older ones, since older people tend to be more concerned about the

33
difficulties they may have in learning new systems (Hertzog & Hultsch, 2000). However, Teo
(2001) found no significant differences in Internet usage for online shopping across age
groups. In terms of Internet usage, differences may exist between the types of content that
individuals seek out depending on age group (Wang & Fesenmaier, 2004b). For instance,
older consumers are less likely to look for new information, whereas younger consumers seek
alternative information and various decision criteria when making purchase decisions
(Evanschitzky & Wunderlich, 2006). Wang and Fesenmaier (2004b) also found that younger
groups (i.e., up to 40 years old) appreciated functional benefits (e.g., information gathering
and ease of transactions) from community participation. Young adults (under age 20) were
more likely to pursue social and psychological benefits (e.g., a sense of belonging and
enhanced social status) than adults over the age of 55. Adults between the ages of 20 and 40
placed more value on hedonic benefits (e.g., entertainment) than other age groups. Likewise,
consumers or members in online communities for hospitality companies (i.e., hotels and
restaurants) may have different reasons to participate in different activities. Based on the age
differences discussed above, the following hypothesis is proposed:
H9: Younger people are more likely to be strongly affected by participation benefits
—functional (H9a), social (H9b), psychological (H9c), hedonic (H9d), and monetary (H9e)
—than are older members of online communities.
Moderating roles of biological gender
Biological gender has been widely used as a moderator variable, particularly in
consumer behavior research (Saad & Gill, 2001). Many studies have shown that social roles
differ based on biological gender differences, which indicate specific behaviors that men or
women are expected to display. For example, men often learn to be assertive and aggressive,
whereas women are more nurturing and tend to be naïve (Putrevu, 2001). These differences
have revealed distinct patterns in communicating and building relationships with others
(Serenko, Turel, & Yol, 2006). For example, men tend to control relationships and dominate
conversations with other people; in contrast, women are more likely to express their personal
feelings, be supportive, and cooperate with others for interaction (Boneva et al., 2006).

34
Similar differences in biological gender have been found in the usage of Websites
(Wasserman & Richmond-Abbott, 2005). Men exhibit preferences for entertainment aspects
such as building Web pages, searching for information about products, and participating in
online games (Weiser, 2000). In contrast, women are more interested in maintaining social
connections through e-mail and online chatting, communicating with friends, and sharing
personal issues and emotions (i.e., social benefits) (Jackson, Ervin, Gardner, & Schmitt,
2001). Nie and Erbring (2000) found that women tend to use email and online chatting more
frequently than men for interpersonal communication (i.e., social benefits). Phillip and Suri
(2004) found that women prefer to receive advertising e-mails more than men do, which
indicates they are less task-oriented (e.g., information search). Based on the above discussion,
differences in biological gender have been observed in online communication and usage
behaviors. Thus, the following hypotheses are proposed:
H10: The gender of online community members moderates the effect of participation
benefits on online community participation.
H10a: The effect of functional benefits on community participation will be stronger
for male members.
H10b: The effect of social benefits on community participation will be stronger for
female members.
H10c: The effect of psychological benefits on community participation will be
stronger for female members.
H10d: The effect of hedonic benefits on community participation will be stronger for
male members.
H10e: The effect of monetary benefits on community participation will be stronger for
female members.
Research Model
Based on the above discussion, the present study proposes a conceptual research
model of relationships: (a) the relationships between community participation and
participation benefits (Figure 2), (b) the relationships between community participation,
brand trust, and brand commitment (Figure 2), and (c) the moderating effect of demographic

35
characteristics (i.e., gender and age) on the relationship between participation benefits and
community (Figure 3 and Figure 4).
Figure 2. Proposed conceptual model for development of an effective online community
CHAPTER 3: RESEARCH METHODOLOGY AND DESIGN
This chapter introduces the research methods utilized to test the H presented in
Chapter 2. The selection of hotel and restaurant brands‘ Facebook pages, sampling and data
collection methods, the survey instrument, and the statistical analysis process are discussed in
the following sections.
Selection of Online Communities in Facebook
Benefit
H1
Social
H2
Benefit H3
Community
Participation
H6 Brand
Hedonic
Benefit H4 H7 H8
Brand
Trust
Monetary
Benefit
H5

36
The present study investigates hotel and restaurant brands‘ Facebook pages. These
two groups were selected because they are the most important segments of the hospitality
industry. Many Facebook pages for hotels and restaurants have incorporated unique features
(e.g., promotions) in order to encourage member participation. Among the numerous fan
pages on the site, four hotel and four restaurant brands‘ Facebook pages were chosen from
the list of ―10 Awesome Hotel Facebook Pages to Like‖
(http://www.businessinsider.com/10- awesome-hotel-facebook-pages-to-like-2011-1) and
―Best Restaurant Facebook Fan Pages‖ (http://hilinskyconsulting.com/blog/2009/11/12/best-
restaurant-facebook-fan-pages/). The former article was published by BusinessInsider.com,
an online community that shares business news. The latter was published by a social media
marketing consulting company. The successfulness of these Facebook pages was based on a
high number of fans as well as a high number of postings by members (Preece et al., 2004).
From the two lists, the following Facebook pages that meet both criteria were chosen: (1)
Marriott Napa Valley Hotel and Spa,
Beacon Hotel, The Westin Dragonara Resort Malta, and The Hermitage Hotel; (2) Outback
Steakhouse, Chili‘s Grill & Bar, Red Lobster, and The Cheesecake Factory.
Sample
The sample for the present study consisted of fans of the hotel and restaurant brands‘
Facebook pages listed above. an online survey was developed and distributed to potential
respondents, both male and female, of at least 18 years of age. The advantages of online
surveys are their (a) low cost, (b) interactivity, (c) high accessibility to the respondent without
time and space constraints, and (d) convenience for data entry and checking (Stopher,
Collins, & Bullock, 2004).
Survey Instrument
The survey consisted of four sections: (1) participation benefits; (2) community
participation, brand trust, and brand commitment; (3) demographic information; and (4)
manipulation checking. Prior to starting the first part of the survey, participants were asked

37
whether they had ever joined either hotel or restaurant brands‘ Facebook pages. Only those
who had confirmed that they have been or currently were a member of a Facebook page
operated by a hospitality company were eligible to complete the survey.
For the hotel questionnaire, a list of hotels indicated in the previous section
(―selection of online communities in Facebook‖) was given as choices for participants to
indicate for which brand pages they are members. For the restaurant questionnaire, a list
of restaurants indicated in the above section (―selection of online communities in
Facebook‖) was given as choices. For respondents who were not a member of given hotel
or restaurant brands‘ Facebook pages, an open-ended question was provided for them to
provide another hotel or restaurant name.
The first part of the survey measured five categories of member benefits (exogenous
variables): functional, social, psychological, hedonic, and monetary benefits, using five-point
Likert-type scales ranging from 1 (not important at all) to 5 (extremely important). Four
benefit variables—functional, social, psychological, and hedonic—were adapted from Wang
and Fesenmaier (2004a); these have been successfully used in a number of studies of online
communities. First, functional benefits were assessed with four items: ―obtaining up-to-date
information‖, ―ease/convenience of communicating with others‖, ―efficiency of online
communication‖, and ―sharing experiences‖. Next, social benefits variables consisted of
four items: ―having trust in the community‖, ―seeking self-identity‖, ―communicating
with other members‖, and ―getting involved with other members‖. Third, three items were
employed to investigate psychological benefits: ―seeking a sense of affiliation in the
community‖, ―seeking a sense of belonging‖, and ―establishing and maintaining
relationships with other members‖. Fourth, hedonic benefits variables consisted of four
items: ―to be entertained by other members‖, ―to have fun‖, ―to seek enjoyment‖, and
―to be entertained‖.
In addition to these four benefit variables, monetary benefits were assessed using
three items adapted from Gwinner et al. (1998) and Lee et al. (2008). These items related to
special deals, discounts, or company events offered on the community site: ―obtaining
discounts or special deals that most consumers don't get‖, ―obtaining better prices than most
consumers‖, and ―receiving free coupons for hotel stays or food/beverages by becoming a
member of this Facebook page‖.

38
The second part of the survey examined levels of community participation, brand
trust, and brand commitment (i.e., endogenous variables). All items in the second part were
measured using a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly
agree). To measure community participation, four items were adapted from Koh and Kim
(2004) and Casaló et al. (2007): ―I take an active part on the hotel/restaurant brand‘s
Facebook page‖, ―I usually provide useful information to other members‖, ―In general, I
post messages and responses on the hotel/restaurant brand‘s Facebook page with great
enthusiasm and frequency‖, and ―I do my best to stimulate the hotel/restaurant brand‘s
Facebook page‖. These items served to gather more detailed information regarding member
behavior than do assessments of use frequency or log-in times (Casaló et al., 2008; Madupu,
2006).
The second part of the survey also included the questions regarding brand trust and
brand commitment. All items for the two constructs were measured using a 5-point Likert-
type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Brand trust was measured
using items from Chiang and Jang‘s (2006) and Wilkins, Merrilees, and
Herington‘s (2010) work. Respondents were asked to rate the extent of their agreement with
the following four statements: ―What the hotel/restaurant brand says about its
products/service is true‖, ―I feel I know what to expect from the hotel/restaurant brand‖,
―the hotel/restaurant brand is very reliable‖, and ―the hotel/restaurant brand meets its
promises‖. To measure brand commitment, three items were adapted from Ahluwalia (2000):
―if the hotel/restaurant brand were not available for reservation (e.g., rooms, tables), it
would make little difference to me if I had to make reservations at other hotels/restaurants‖,
―I consider myself to be highly loyal to the hotel/restaurant brand‖, and ―when another
brand has a special deal (e.g., lower room rate/price for meal), I generally stay at the hotel/
visit the restaurant with the better deal‖.
The third part of the survey elicited demographic information such as education,
biological gender, and age. Age ranges were adapted from Wang and Fesenmaier‘s (2004a):
(1) younger than 21, (2) between 21 and 30, (3) between 31 and 40, (4) between 41 and 55,
and (5) over 55. Several open-ended questions were also included in this part of the survey:
(1) How long have you been a member of this hotel's/restaurant's Facebook page?, (2) How

39
long, on average, do you participate in this brand‘s hotel/restaurant Facebook page each
week?, and (3) How many Facebook pages of hotels/restaurants are you a member of?
In the last part of the questionnaire, respondents were asked about the perceived
success of the Facebook page as a manipulation check: ―The interaction between the
company and other members is active‖, ―The hotel/restaurant brand‘s Facebook page is
successful‖, and ―I like visiting the hotel/restaurant brand‘s Facebook page‖. All items were
measured using a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly
agree).
A pilot test was undertaken by distributing the survey to a total of 15 graduate
students majoring in hotel management to determine whether wording changes were needed
to enhance clarity. Based on the feedback gathered, minor changes were made to ensure that
participants would have no difficulty understanding or answering questions.
Table 5. Constructs and items of the survey
Functiona  Obtaining up-to-date information l benefits 
Ease/convenience of communicating with others
• Efficiency of online communication
• Sharing experiences
Social benefits  Having trust in the community
• Seeking self-identity
• Communicating with other members
• Getting involved with other members
Psychologic
al benefits
• Seeking a sense of affiliation in the community
• Seeking a sense of belonging
• Establishing and maintaining relationships with other members
Hedoni
c
benefits
• To be entertained by other members
• To have fun
• To seek enjoyment
• To be entertained
Monetar
y
benefits
• Obtaining discounts or special deals that most consumers don't get
• Obtaining better prices than most consumers
Construct Measurement items

40
• Receiving free coupons for hotel stays or food/beverages by becoming a
member of this community
Communit
y
participatio
n
• I take an active part in the hotel (restaurant) brand‘s Facebook
page
• I usually provide useful information to other members in the hotel
(restaurant) brand‘s Facebook page
• In general, I post messages and responses in the hotel (restaurant)
brand‘s Facebook page with great enthusiasm and frequency  I do my
best to stimulate the hotel (restaurant) brand.
Brand trust  What the hotel (restaurant) brand says about its products/service is true
• I feel I know what to expect from the hotel (restaurant) brand.
• The hotel (restaurant) brand is very reliable.
• The hotel (restaurant) brand meets its promises.
Brand  If the hotel (restaurant) brand were not available for reservations (e.g.,
commitment rooms, tables), it would make little difference to me if I had to make
reservations at other hotel/restaurant brand.*
• I consider myself to be highly loyal to the hotel (restaurant) brand.
• When another brand has a special deal (e.g., lower room rate/price for
meal), I generally stay at the hotel/visit the restaurant with the better deal
rather than the hotel (restaurant) brand.*
Note. Reverse-coded item*

41
Data Collection
The data were collected between June 25 and July 5, 2011. Participants were
recruited from two sources. First, a panel of participants identified by the online research
company, Qualtrics, was used.. The data on the fans of hotel brands‘ Facebook pages were
collected from both the Qualtrics panel (154 responses) and university alumni (60
responses); whereas the data on the fans of restaurant brands‘ Facebook pages were collected
from university alumni.
An email invitation was sent to potential participants, along with a link to the online
questionnaire. The invitation sent by Qualtrics included a $1 incentive for each of their panel
members,. A total of 21,000 invitations were sent to the alumni list. From the 21,000 alumni,
452 responses were received (60 from members of hotel brands‘ Facebook pages and 392
from members of restaurant brands‘ Facebook pages), with the response rate of 2.15%.
Because of the low response rate in the category of hotel respondents, the present researcher
determined to employ Qualitrics, an online research company, to further collect data from
fans of hotel brands‘ Facebook pages. A total of 5,000 invitations were sent to the panelists
of the company‘s database. The response rate was 3.08%; 154 responses were collected.
Data Analysis
In the data analysis process, descriptive statistics, including frequencies and
percentages, were used for demographic data. Furthermore, the mean values for each item
were calculated.
The present study employed the two-step structural equation modeling (SEM)
approach suggested by Anderson and Gerbing (1988). The first step involved confirmatory
factor analysis (CFA) to validate the scales for the measurement of specific constructs
proposed in the research model (Hair, Anderson, Tatham, & Black, 1998). When using CFA,
items that produce factor loadings lower than 0.5, the cut-off value suggested by Hair, Black,
Babin, Anderson, and Tatham (2006), were deleted. The second step involved examination of
the structural model through SEM in order to evaluate the validity of the proposed model and
H. The maximum likelihood procedure was used to estimate the measurement model and
structural model (Namkung & Jang, 2007) in Amos 6.0.

42
Measurement model
CFA was utilized to evaluate the overall measurement quality (Anderson & Gerbing,
1992), while a reliability test (Cronbach‘s alpha) was conducted to assess the internal
consistency of each construct. The cutoff value of .70 for Cronbach‘s alpha (Nunnally, 1978)
was used. A significant conventional chi-square test (χ2) statistic indicated a poor fit. The
cutoff point of χ2/df was set at 3:1 (Joreskog & Sorbom, 1988). In other words, if the ratio
(χ2/df) fell between 1 and 3, the model fit was perceived as acceptable (McIver & Carmines,
1981). TLI and CFI values greater than .90 indicated a satisfactory model fit (Hair et al.,
2006; Yuan & Jang, 2008; He & Song, 2009). These two indices can be influenced by the
average size of the correlations in the data. If the average correlation between variables is
low, then the TLI (and the CFI) will have a low score (Kenny, 2010). RMSEA with a value
below .08 was recommended (Byrne, 1998; Diamantopoulos & Siguaw, 2000).
Structural model
Two structural models were tested. The first assessed the proposed causal
relationships between participation benefits (functional, psychological, social, hedonic, and
monetary), community participation, brand trust, and brand commitment (Figure 2.1) which
reflected H1 through H8. The second model examined the moderating role of two
demographic variables, age (Figure 2.2; H9a~e) and biological gender (Figure 2.3; H10a~d),
on the relationships between participation benefits and community participation using a
multi-group SEM approach suggested by Joreskog and Sorbom (1993). The mediating effect
of brand trust between community participation and brand commitment was tested using
Baron and Kenny‘s approach (1986).
CHAPTER 4: RESULTS
This chapter reports the results of the analysis, which include demographic
characteristics of the sample, descriptive statistics of the variables, and measurement and
structural equation model tests.

43
Demographic Characteristics
A total of 606 questionnaires were collected (214 responses from hotel respondents
and 392 responses from restaurant respondents). The data were screened to control response
bias. Responses that included one or more unanswered sections and those with extreme
answers were removed. After deleting 72 invalid surveys, 534 responses (203 hotel responses
and 331 restaurant responses) were kept for further analysis. Table 6 and Table 7 present the
demographic profile of the hotel and the restaurant respondents, respectively. In addition to
demographic characteristics such as biological gender, age, and education, tables include
respondents‘ information regarding geographic regions where respondents reside, duration of
membership, average time spent on hotel or restaurant brands‘ Facebook pages per week, and
the number of Facebook page memberships.
In the hotel study (Table 6), approximately 52% of the participants were female and
48% were male. Among them, 54% ranged in age from 21 to 40 years old. 33.2% of the
participants had completed bachelor degrees and 29.5% had earned a graduate degree. The
majority of participants (97.4%) were Americans. Approximately 74% of the participants had
been members of hotel brands‘ Facebook pages for less than 12 months. In terms of the usage
of hotel brands‘ Facebook pages, over half of the participants had spent one to five hours per
week (65.6 %) on the pages, and belonged to two to five hotel brands‘ Facebook pages
(55.4%).
For the restaurant sample (Table 7), the majority of respondents were female; 70.3%
of the participants were female and 29.7% were male. The majority of restaurant participants‘
ages ranged from 21 to 30 (53.4%), followed by 31-40 age group (25.5%). The majority of
participants (97.9%) was American and highly educated (80.2%); 38.3% of respondents had
completed a bachelor degree, while 41.9% possessed a graduate degree.
Sixty-five percent of the participants had been members of restaurant brands‘ Facebook pages
for less than a year. In terms of the usage of restaurant brands‘ Facebook page, more than a
half of the participants had spent one to five hours per week (56.4%) and belonged to two to
five restaurant brands‘ Facebook pages (69.1%).

44
Table 6. Demographic characteristics of the hotel sample
Demographic characteristics Frequenc Percentage
y
Biological gender (n = 195)
Male 101 51.8
Female 94 48.2
Age (n = 193)
18-20 years old 10 5.2
21-30 59 30.6
31-40 45 23.3
41-55 54 28.0
Over
55
25 13.0
Education (n = 190)
High school or less 35 18.4
Associate degree 36 18.9
Bachelor degree 63 33.2
Graduate degree 56 29.5
Geographic region (n = 193)
United State of America 188 97.4
European 2 1.0
Asian 3 1.6
Duration of membership
(n = 193)
Less than 12 months 142 73.6
12-24 months 41 21.2
25-36 months 6 3.1

45
Over 36 months 4 2.1
Average hour spent per week
on Facebook pages (n = 192)
Less than 1 hour 46 24.0
1- 5 hours 126 65.6
6-10 hours 9 4.7
More than 10 hours 11 5.7
Number of Facebook page
memberships (n = 193)
1 membership 39 20.2
2-5 memberships 107 55.4
6-10 memberships 21 10.9
More than 10 memberships 26 13.5

46
Table 7. Demographic characteristics of the restaurant sample
Demographic characteristics Frequenc Percentage
y
Biological gender (n = 327)
Male 97 29.7
Female 230 70.3
Age (n = 326)
18-20 years old 36 11.0
21-30 174 53.4
31-40 83 25.5
41-55 29 8.9
Over
55
4 1.2
Education (n = 329)
High school or less 49 14.9
Associate degree 15 4.6
Bachelor degree 126 38.3
Graduate degree 138 41.9
Geographic region (n = 330)
United State of America 323 97.9
European 2 0.6
Asian 5 1.5
Duration of membership
(n = 324)
Less than 12 months 212 65.4
12-24 months 84 25.9
25-36 months 28 8.6

47
Over 36 months 0 0.0
Average hours spend per week on
Facebook pages (n = 328)
Less than 1 hour 185 56.4
1- 5 hours 99 30.2
6-10 hours 34 10.4
More than 10 hours 10 3.0
Number of Facebook page
memberships (n = 320)
1 membership 56 17.5
2-5 memberships 221 69.1
6-10 memberships 18 5.6
More than 10 memberships 25 7.8

48
Brand Profile and Manipulation Check
Table 8 shows the profiles of hotel and restaurant brands listed in the previous section
(―selection of online communities in Facebook‖). More than 50% of the hotel data were
from the Marriott hotel group and 32% were from 36 different hotel brands including
Hilton, Intercontinental, Ritz Carlton, and Westin. On the other hand, approximately 54%
of the restaurant data were from 135 different brands including Olive Garden, Panera,
Texas Roadhouse, and Chipotle Mexican Grill.
To check the successfulness of Facebook pages operated by the hospitality companies
named by the respondents, the mean values for the three items related to the perceived
successfulness of the Facebook page were calculated for each brand (Table 9). All of the
mean values were above 3.0, which indicated that participants generally perceived the
Facebook pages to be actively managed, successful, and they liked visiting the brands‘ pages.
Accordingly, the respondents created an appropriate sample for the present study due to their
strong interest and concern for Facebook page brands related to the hotel and restaurant
establishments.
Table 8. Brand profile of the sample
Hotel (n = 203) Frequenc
y
Percen
t
Marriott 108 53.2
Stanley 10 4.9
Beacon 7 3.5
The Algonquin 13 6.4
Others (specified by respondents). 65 32.0
Restaurant (n = 331) Frequenc
y
Percen
t
Outback Steakhouse 28 8.5
Chili‘s 32 9.7

49
Red Lobster 32 9.7
The Cheesecake Factory 61 18.4
Other (specified by respondents). 178 53.8
Table 9. Perceived success of Facebook pages
Hotel (n = 203) Mea
n
SD Min. Max
.
There is active participation between the hotel
and members.
3.72 .85 2 5
The hotel brand's Facebook page is successful. 3.96 .80 1 5
I like visiting the hotel brand's Facebook page. 3.91 .80 1 5
Restaurant (n = 331) Mea
n
SD Min. Max
.
There is active participation between the
restaurant and members.
3.36 .89 1 5
The restaurant brand's Facebook page is 3.76 .67 1 5

50
64
successful.
I like visiting the restaurant brand's Facebook 3.52 .76 1 5 page.

Descriptive Statistics for Measures
Table 10 reports the descriptive statistics of the hotel and restaurant studies, including
empirical items for each construct, mean, standard deviation, and minimum and maximum of
each measurement item. These statistics were used to understand the variation of each item
for the proposed constructs measured in the causal model. The constructs were functional
benefit, social benefit, psychological benefit, hedonic benefit, monetary benefit,
participation, brand trust, and brand commitment.

Table 10. Descriptive statistics for all items used to measure model constructs
Constructs
Items
Hotel (n=203) Restaurant (n=331)
Mean SD Min. Max. Mean SD Min. Max.
Functional benefit
Obtaining up-to-date information about the
Hotel (Restaurant) brand
4.22 .84 1 5 3.84 .89 1 5
Conveniently communicating with others online 3.82 .98 1 5 3.14 1.08 1 5
Efficiently communicating online 3.95 .90 1 5 3.45 1.03 1 5
Sharing experiences in the Hotel (Restaurant)
brand
4.07 .89 1 5 3.72 .97 1 5
Social benefit
Having trust in the community of Facebook 4.03 .91 1 5 3.43 1.04 1 5
Seeking self-identity 3.40 1.15 1 5 2.68 1.10 1 5
Communicating with other members 3.82 1.00 1 5 3.10 1.06 1 5
Getting involved with other members 3.52 1.00 1 5 2.84 1.03 1 5
Psychological benefit
Seeking a sense of affiliation in the community 3.53 .99 1 5 3.09 1.11 1 5
Seeking a sense of belonging 3.53 1.07 1 5 2.87 1.07 1 5

Table 10. (continued)
Establishing and maintaining relationships with
other members of Facebook
3.72 1.07 1 5 3.03 1.11 1 5
Being amused by other members 3.64 .93 1 5 3.37 1.01 1 5
Having fun on the brand's Facebook page 3.81 .91 1 5 3.33 1.03 1 5
Seeking enjoyment on this Facebook page 3.79 .91 1 5 3.24 1.06 1 5
Being entertained on this Facebook page 3.77 .91 1 5 3.34 1.00 1 5
Hedonic benefit
Table 10.
Descriptive
statistics for all
items used to
measure model
constructs
Note. Reverse-coded item*
63

Hotel (n=203) Restaurant (n=331)
Mean SD Min. Max. Mean SD Min. Max.
Monetary benefit
Obtaining discounts or special deals that most
consumers don't get
4.44 .75 1 5 4.53 .75 1 5
Obtaining better prices than other consumers 4.36 .82 1 5 4.28 .85 1 5
Receiving free coupons for the Hotel
(Restaurant) brand by becoming a member of
the Facebook page
4.32 .83 1 5 4.42 .81 1 5
Participation
I take an active part in the Hotel (Restaurant)
brand's Facebook page
3.27 1.07 1 5 2.45 .97 1
5
I frequently provide useful information to other
members
3.35 1.03 1 5 2.14 .87 1 5
In general, I post messages and responses on the
brand's Facebook page with great enthusiasm
and frequency
3.16 1.09 1 5 2.12 .92 1 5
I do my best to participate in activities offered
on the brand's Facebook page
3.43 1.02 1 5 2.53 .100 1 5
Brand trust
What the Hotel (Restaurant) brand says about 3.91 .78 2 5 3.72 .77 1 5
64

Table 10. (continued)
its products/service is true
I feel I know what to expect from the Hotel 4.06 .76 1 5 3.98 .62 1 5
(Restaurant) brand
The Hotel (Restaurant) brand is very reliable 4.15 .67 2 5 3.97 .65 1 5 The Hotel
(Restaurant) brand meets its 4.11 .74 1 5 3.98 .630 1 5 promises

Table 10. (continued)
Hotel (n=203) Restaurant (n=331)
Mean SD Min. Max. Mean SD Min. Max.
Brand commitment
If the Hotel (Restaurant) brand had no
available reservations, I would have no
problem finding a different Hotel (Restaurant)
with which I would want to make
reservations*
2.32 .99 1 5 2.23 .92 1
5
I consider myself to be highly loyal to the
Hotel (Restaurant) brand
3.65 .99 1 5 3.22 .99 1 5

Table 10. (continued)
When another brand has a special deal (e.g.,
discounted room rate/ discount price for meal),
I generally visit that Hotel (Restaurant) with
the better deal*
2.43 1.08 1 5 2.36 .98 1 5
65

66
Measurement Model
The measurement model consisted of eight latent variables: five benefit variables
(functional, social, psychological, hedonic, and monetary benefits), participation, brand trust,
and brand commitment. Confirmatory factor analysis was conducted to evaluate the overall
fit of measurement items in the conceptual model.
Confirmatory factor analysis (CFA) for the hotel study
The initial measurement model of the hotel brands‘ Facebook pages was comprised
of 29 measurement items. The initial estimation of this measurement model did not fit well.
The chi-square value of 796.11 with 349 degrees of freedom was statistically significant at
p
< .001. Moreover, the other model fit indices used in the study were not acceptable (TLI
= .84, CFI = .86, RMSEA = .08). Based on the results of confirmatory factor analysis (CFA),
one measurement item of brand commitment, ―I consider myself to be highly loyal to the
hotel brand.‖ was deleted because it presented a factor loading lower than 0.50 (i.e., .39)
which is the cut-off value suggested by Hair et al. (2006).
After deleting this item, CFA was conducted with the 28 measurement items, and the
model fit for the revised measurement model was still found to be unacceptable (χ² = 691.23,
df = 322, p < .001, TLI=.86, CFI=.88, RMSEA=.075). Due to the poor model fit, the present
researcher checked the correlation coefficients between the variables of the causal model
(Table 11). The results showed that the exogenous variables (functional, social,
psychological, and hedonic benefits) were highly correlated with each other, with correlations
ranging from .53 to .87. This indicated multi-collinearity problems among the exogenous
variables. To resolve this issue, the present researcher carried out two processes in order to
eliminate highly correlated items:
• Identified constructs to be compounded as a single construct;

67
• Deleted items that produced the lowest factor loadings from highly correlated
constructs, even when all factor loadings were equal to or greater than .50.
First, social and psychological benefits were combined into a single construct due to
their high correlation (r =.87). This compound construct was renamed as ‗social-
psychological benefits‘ for later CFA. Second, measurement items with the lowest factor
loadings were identified from the four exogenous variables that were highly correlated with
each other and a total of four items were eliminated for a better model fit. These items were:
―obtaining up-to-date information about the hotel brand (.56)‖, ―having trust in the
Facebook community (.57), ―establishing and maintaining relationships with other members
of
Facebook (.75)‖, and ―being amused by other members (.66).‖
After deleting these four measurement items, CFA was conducted with 24 items for
the seven latent constructs (functional benefits, social-psychological benefits, hedonic
benefits, monetary benefits, participation, brand trust, and brand commitment). The fit for the
measurement model with two revisions was still not acceptable at χ² = 515.23, df = 231, p
< .001, TLI = .87, CFI = .89, RMSEA = .078. Using the same process, social-psychological
benefits were found to be highly correlated with functional benefits (r = .67) and hedonic
benefits (r = .64). The two items of social-psychological benefits that were primarily
responsible for the multi-collinearity problem were deleted. These two items were ―seeking
self-identity (.70)‖ and ―communicating with other members (.72)‖.
Table 11. Correlation coefficients of constructs: initial measurement model for the hotel
study
Constructs 1 2 3 4 5 6 7 8
1. Functional
benefits 1
2. Social
benefits .77** 1
3. Psychological
benefits .62** .87** 1
4. Hedonic
benefits .53** .73** .59** 1

68
5. Monetary
benefits .50** .31** .26** .26** 1
6. Participation .50** .60** .53** .49** .13 1
7. Brand trust .38** .30** .24** .27** .49** .45** 1
8. Brand
commitment -.18 -.17 -.21** -.07 -.03 -.02 .09 1
Note: non-significant; *p < .05; **p < .01
After deleting two measurement items of social-psychological benefits, the final
model, consisting of seven latent variables with 22 items, was tested. All the variables
included at least three measurement items, with the exception of brand commitment. The
CFA results showed a satisfactory model fit (χ² = 355.22, df = 188, p < .001, TLI = .91, CFI
= .93, RMSEA = .066). Since the ratio (χ2/df = 1.89) fell between 1 and 3, the model fit was
determined to be acceptable (McIver & Carmines, 1981). The values for TLI and CFI were
greater than .90 and the value for RMSEA was below .08, indicating a satisfactory model fit
(Byrne, 1998; Diamantopoulos & Siguaw, 2000; Hair et al., 2006). The correlation
coefficients among the variables are illustrated in Table 12. All variables in the final model
(functional benefits, social-psychological benefits, hedonic benefits, monetary benefits,
participation, brand trust, and brand commitment) were either moderately or highly
correlated with each other, with the correlations ranging from -.19 to .59. Table 13 shows
final measurement items with factor loadings and Cronbach‘s alpha estimates for the
constructs. All the factor loadings in the final measurement model were equal to or greater
than .59. The Cronbach‘s alpha estimates for the constructs in the present study ranged
from .79 to .88, which were above the cutoff value of .70 (Hair et al., 1998; Nunnally, 1978).
Thus, the data showed an acceptable level of internal consistency.

69
Table 12. Correlation coefficients of constructs: final measurement model for the hotel
study
Constructs 1 3 4 5 6 7 8
1. Functional
benefits 1
3. Social-
Psychological
benefits
.59** 1
4. Hedonic
benefits .49** .57** 1
5. Monetary
benefits .44** .28** .24** 1
6. Participation .51** .52** .48** .13 1
7. Brand trust .34** .25** .27** .49** .45** 1
8. Brand
commitment -.17 -.19** -.06 -.03 -.02 .09 1
Note: non-significant; *p < .05; **p < .01
Table 13. Item measurement properties for the hotel study Construct
Standardized Cronbach’s
Factor Loadings alpha
Functional benefits .80
Obtaining up-to-date information about the hotel
brand. d
-
Efficiently communicating with others online. .84
Conveniently communicating online. .86
Sharing experiences in the hotel brand. .59
Social-Psychological benefits .86
Having trust in the Facebook community. d -

70
Seeking self-identity. d -
Communicating with other members. d -
Getting involved with other members. .72
Seeking a sense of affiliation in the community. .87
Seeking a sense of belonging. .88
Establishing and maintaining relationships with other
members of Facebook. d -
Hedonic benefits .85
Being amused by other members. d -
Having fun. .77
Seeking enjoyment. .90
Being entertained. .75
Monetary benefits .85
Obtaining discounts or special deals that most
consumers don't get.
.77
Obtaining better prices than other consumers. .88
Receiving free coupons for the hotel brand by becoming .78
a member of the Facebook page.
Note: d item removed from the original scale
Table 13. (continued)
Standardized
Factor Loadings
Cronbach’s
alpha
Participation .88
I take an active part in the hotel brand‘s Facebook page. .83
I frequently provide useful information to other
members.
.81
In general, I post messages and responses on the
brand‘s Facebook page with great enthusiasm and
frequency.
.81

71
I do my best to participate in activities offered on the
brand‘s Facebook page.
.76
Brand trust .84
What the hotel brand says about its products/service is
true.
.66
I feel I know what to expect from the hotel brand. .78
The hotel brand is very reliable. .81
The hotel brand meets its promises. .79
Brand commitment .79
If the hotel brand had no available reservations, I would
have no problem finding a different hotel with which I
would want to make reservations.
.60
I consider myself to be highly loyal to the hotel brand. d -
When another brand has a special deal (e.g., discounted
room rates), I generally visit the hotel with the better
deal.
1.09
Confirmatory factor analysis (CFA) for the restaurant study
The initial measurement model of the restaurant brands‘ Facebook pages was
comprised of 29 measurement items. The initial measurement estimation of this model did
not fit well (χ² = 947.62, df = 349, p < .001, TLI = .87, CFI = .89, RMSEA = .072). Three
measurement items were found to have factor loadings lower than the cutoff value of .5
(Hair et al., 2006). These items were ―obtaining up-to-date information about the
restaurant brand (.27)‖, ―sharing experiences in the restaurant brand (.49)‖, and ―having
trust in the
Facebook community (.48)‖. To keep at least three measurement items in the exogenous
construct, the item with the factor loading of .49 was retained. Thus, two measurement
items were removed based on the factor loadings.

72
After the deletions were made, CFA was conducted with the 27 measurement items.
The model fit for the second measurement model was not acceptable at χ² = 770.77, df = 296,
p < .001, TLI = .89, CFI = .91, RMSEA = .070. Since the model fit was low, correlation
coefficients between the variables of the causal model were investigated. Table 14 illustrates
the correlation coefficients among the variables. Similar to the results obtained with the hotel
study, multi-collinearity problems among exogenous variables were detected, with the
correlations ranging from .51 to .94.
The same CFA process utilized for the hotel study was conducted in the restaurant
study. Due to the extremely high correlation of the social and psychological benefit
constructs (r = .94), these two constructs were combined into a single construct, which was
re-named as ―social-psychological benefits‖ for later CFA. All the factor loadings were
equal to or greater than .52, with the exception of an item of functional benefits (―sharing
experiences in the restaurant brand‖ = .49). Due to the high correlation among the three
constructs (social, psychology, and hedonic benefits), three measurement items (one for each
of the three constructs) were deleted even though their factor loadings were acceptable. These
items were ―seeking self-identity (.70)‖, ―establishing and maintaining relationships with
other members of Facebook (.69)‖, and ―being amused by other members (.52)‖.
After deleting the three measurement items above, CFA was conducted with 24 items
for seven latent constructs (functional benefits, social-psychological benefits, hedonic
benefits, monetary benefits, participation, brand trust, and brand commitment). The fit for the
measurement model with two revisions was satisfactory at χ² = 526.34, df = 231, p < .001,
TLI = .92, CFI = .93, RMSEA = .062. Although the model fit was acceptable, the correlation
coefficients between functional and social-psychological benefits were still high (r = .62).
Therefore, the measurement item that produced the lowest factor loading of social-
psychological benefits was deleted (―communicating with other members‖ = .79). In
addition, to keep the same measurement items as the causal model in the hotel study, the
present researcher determined to remove one measurement item of brand commitment that
had been deleted in the hotel response sample, although the factor loading of this item
was .74. The item was ―I consider myself to be highly loyal to the restaurant brand‖.
Table 14. Correlation coefficients of constructs: initial measurement model for the
restaurant study
Constructs 1 2 3 4 5 6 7 8

73
1. Functional
benefits 1
2. Social
benefits
.63**
1
3. Psychological
benefits
.58** .94**
1
4. Hedonic
benefits
.44** .51** .52**
1
5. Monetary
benefits
.02 -.10 -.08 .12
1
6. Participation .28** .41** .42** .33** -.06 1
7. Brand trust .18** .19** .29** .21** .06 .19** 1
8. Brand
commitment
.13 .17 .21 .03 -.27** .29** .45** 1
Note: non-significant; *p < .05; **p < .01
After deleting the two measurement items above, the final model, consisting of
seven latent variables with 22 items, was tested. All variables included at least three
measurement items, with the exception of brand commitment. The CFA results showed a
satisfactory model fit (χ² = 337.03, df = 188, p < .001, TLI = .95, CFI = .96, RMSEA
= .049). Based on the ratio (χ2/df = 1.79), the model fit was perceived as acceptable because
the ratio fell between 1 and 3 (McIver & Carmines, 1981). The values for TLI and CFI were
greater than .90 and the value for RMSEA was below .08, which indicated a satisfactory
model fit (Byrne, 1998; Diamantopoulos & Siguaw, 2000; Hair et al., 2006). Table 15
presents the correlation coefficients among the variables. All variables in the final model
(functional benefits, social-psychological benefits, hedonic benefits, monetary benefits,
participation, brand trust, and brand commitment) were moderately to highly correlated with
each other, with correlations ranging from -.40 to .55. Table 16 shows the final
measurement items with factor loadings and Cronbach‘s alpha estimates for each construct.
All factor loadings in the final measurement model were equal to or greater than .53.

74
Cronbach‘s alpha estimates for the constructs in the present study ranged from .75 to .89,
with the exception of the brand commitment variable. All the Cronbach‘s alpha values were
greater than .70, indicating a good level of internal consistency (Nunnally, 1978). The
Cronbach‘s alpha estimate for brand commitment (.60) was also acceptable, being at or
above .60 (Horne, Hankin, & Jenkins, 2001; Nully, 1967; Ogilvie et al., 2007). Therefore,
the data showed an acceptable level of internal consistency.
Table 15. Correlation coefficients of constructs: final measurement model for the
restaurant study
Constructs 1 2 3 4 5 6 7
1. Functional
benefits 1
3. Social-
Psychological
benefits
.55** 1
4. Hedonic
benefits
.43** .49**
.1
5. Monetary
benefits
.03 -.10 .12*
1
6. Participation .27** .39** .33** -.06 1
7. Brand trust .18** .28** .20** .06 .19** 1
8. Brand commitment .07 .14 -.09 -.40** .16** .27** 1
Note: non-significant; *p < .05; **p < .01
Table 16. Item measurement properties for the restaurant study Construct
Standardized Cronbach’s
Factor Loadings alpha
Functional benefits .75

75
Obtaining up-to-date information about the restaurant
brand. d
-
Efficiently communicating with others online. .77
Conveniently communicating online. .90
Sharing experiences about the restaurant brand. .49
Social-Psychological benefits .85
Having trust in the Facebook community. d -
Seeking self-identity. d -
Communicating with other members. d -
Getting involved with other members. .72
Seeking a sense of affiliation in the community. .87
Seeking a sense of belonging. .86
Establishing and maintaining relationships with other
members of Facebook. d -
Hedonic benefits . 88
Being amused by other members. d -
Having fun. .80
Seeking enjoyment. .85
Being entertained. .87
Monetary benefits .88
Obtaining discounts or special deals that most consumers
don't get.
.92
Obtaining better prices than other consumers. .84
Receiving free coupons for the restaurant brand by .78
becoming a member of the Facebook page.
Note: d Item removed from the original scale.
Table 16. (continued)
Standardized
Factor Loadings
Cronbach’s
alpha

76
Participation .87
I take an active part in the restaurant‘s Facebook page. .74
I frequently provide useful information to other
members.
.85
In general, I post messages and responses on the
brand‘s Facebook page with great enthusiasm and
frequency.
.86
I do my best to participate in activities offered on the
brand‘s Facebook page.
.71
Brand trust .89
What the restaurant brand says about its
products/service is true.
.63
I feel I know what to expect from the restaurant brand. .83
The restaurant brand is very reliable. .95
The restaurant brand meets its promises. .90
Brand commitment .60
If the restaurant brand had no available reservations, I
would have no problem finding a different restaurant
with which I would want to make reservations.
.53
I consider myself to be highly loyal to the restaurant
brand. d
-
When another brand has a special deal (e.g., discount
price for meal), I generally visit the restaurant with the
better deal.
.81
Structural Model
Proposed the causal relationships among five exogenous (functional, social, psychological,
hedonic, and monetary benefits) and three endogenous (participation, brand trust, and brand
commitment) constructs. A structural equation model was estimated using a

77
maximumlikelihood estimation procedure. The two figures provide standardized path
coefficients (β) and t-values for each significant path of the conceptual model.
Testing the structural model for the hotel study
Concentrating on the proposed causal relationships derived from the hypotheses. Since
social and psychological constructs were combined into a single construct
(socialpsychological benefits), H2 and H3 were deleted. A new path between social-
psychological benefits and participation was indicated as H11: social-psychological benefits
have a positive influence on community participation. All indices illustrated a satisfactory
model fit (χ² = 403.97, df = 196, p < .001, CFI = .91, RMSEA = .072) with the exception of
TLI (.90). The chi-square ratio (χ2/df) was 2.06, which was acceptable.
Among the seven paths proposed in the conceptual model, only four paths were
statistically significant: the path from functional benefits to participation (β = .31, t = 2.91, p
< .05), the path from social-psychological benefits to participation (β = .24, t = 2.42, p < .05),
the path from hedonic benefits to participation (β = .23, t = 2.51, p < .05), and the path from
participation to brand trust (β = .46, t = 5.67, p < .001). These results statistically supported

81
H1, H4, H7, and H11. In other words, consumers‘ participation benefits (functional,
hedonic, and social-psychological benefits) positively influence consumer participation, and
this participation has a significant influence on brand trust.
Three hypotheses were not supported: H5, which predicted a positive effect of
monetary benefits on consumer participation; H6, which posited the positive effect of
participation on brand commitment; and H8, which posited the positive effect of brand trust
on brand commitment. Due to the rejection of H6 and H8, the mediating effect of brand trust
on the relationship between participation and brand commitment was not tested. The
summary of this causal model is illustrated in Table 17
Table 17. Summary of support for hypotheses based on the results of SEM in the
conceptual model (hotel study)
Hypothesis Path Proposed
effect
Result
H1 Functional benefits→ Participation + s.
H4 Hedonic benefits→ Participation + s.
H5 Monetary benefits→ Participation + n.
H6 Participation→ Brand commitment + n.
H7 Participation→ Brand trust + s.
H8 Brand trust→ Brand commitment n.
H11 Social-Psychological benefits→ Participation + s.
Note: n.= non-significant; s. = significant
Testing the fully recursive model for the hotel study
A fully recursive model including all the plausible paths was constructed and
estimated. The model generated a total of 15 paths, with 8 paths more than the original
conceptual model. The fully recursive model was significant at χ² = 355.22, df = 188, p <
.001. The model fit was also satisfactory (TLI = .91, CFI = .93, RMSEA

82
= .066). Since the chi-square ratio (χ2/df) was 1.89, which fell between 1 and 3, the model fit
was perceived as acceptable (McIver & Carmines, 1981). The χ² values of the fully recursive
model decreased to 48.75 with 8 df, which was statistically significant at p < .001. In
comparison with the finalized conceptual model, the fully recursive model indicated a better
fit, according to the goodness-of-fit indicators. From the results, the fully recursive model
appeared to be more suitable than the conceptual model (Table 18).

83
The significant paths were the same as the conceptual model. By testing the fully
recursive model, the present study identified a new, direct path from monetary benefits to
brand trust (H12a). The standardized path coefficient between monetary benefits and brand
trust was .48, which was statistically significant (t = 5.36, p < .001). This result indicated
that monetary benefits have a significant influence on brand trust. Although the relationship
between the two constructs was not proposed, the structural model with this additional path
indicated significantly improved model fit indices. Table 19 shows path coefficients and t-
values for each path in the reduced (theoretical) model and the fully recursive model. The
new path will be discussed in chapter 5.
Table 18. Chi-square test of model comparison for the hotel study
Model comparison 2 χ df χ2/df TLI CFA RMSEA
Conceptual model 403.97 196 2.06 .90 .91 .07
Fully recursive model 355.22 188 1.89 . 91 .93 .066
Δχ2(df) 48.75 (8)
p < .001

Table 19. Unstandardized path coefficients and t-Values for structural model (hotel study)
Reduced (theoretical) model Fully recursive model
Community
participation
Brand
trust
Brand
commitment
Community
participation
Brand trust Brand
commitment
Functional
benefit
.52(2.91)
b(t) b(t)
.52(2.91) -.05(-.49) -.19(-1.36)
Social-psychological
benefit
.25(2.42) .25(2.49) -.06(-.98) -.15(-1.63)
Hedonic
benefit
.25(2.51) .24(2.48) .01(.19) .05(.77)
Monetary
benefit
-.14(-1.32) .-18(-1.68) .40(5.36) .00(.04)
Community
participation
.28(5.67) .12(1.37) .28(5.36) .17(1.35)
Brand
trust
-.27(-1.66) .05(.65)
R2 .38 .21 .00 .38 .41 .08
Model fit χ² = 403.97, df = 196, TLI = .90,
CFI = .91, RMSEA = .072
χ² = 355.22, df = 188, TLI = .91,
CFI = .93, RMSEA = .066
83

85
Testing for moderating effects of age and biological gender for the hotel study
The moderating effects of age and biological gender were estimated through a multi-
group analysis process proposed by Joreskog and Sorbom (1988). In the conceptual model,
the moderating effects of age and biological gender on the paths between participation
benefits (functional, social-psychological, hedonic, and monetary benefits) and community
participation were examined. Since social and psychological benefits were combined into
one construct, H9b-c and H10b-c were removed from the moderating model. The
relationship between social-psychological benefits and community participation generated a
new path. Thus, the present researcher proposed new hypotheses in regard to the effects of
age (H9f) and biological gender (H10f) on the relationships between social-psychological
benefits and community participation.
The moderating effects were tested in two procedures. First, a chi-square difference
test was conducted between a constrained and an unconstrained model. The constrained
model set all the paths, variances of latent variables, and factor loadings to be equal across
the moderating groups, whereas the unconstrained model released all the paths that were
restricted in the constrained model. Second, the constrained model was re-estimated by
releasing the restriction of equal path estimates for one particular path. Since this model had
one degree of freedom less than the model with all constrained paths, a significant model
improvement was achieved when the drop in χ² between the two models for one degree of
freedom was higher than 3.84 (p < 0.05). These procedures were used for testing the
moderating effects of both age and biological gender.
To test the moderating effect of age, the present researcher used a median-split
procedure to create elder and younger age groups (Harrington, Ottenbacher, & Kendall,
2011; Park, Yang, & Lehto, 2007). According to the median of age, participants who
indicated their age as under or equal to 40 years were assigned to the younger group,

86
whereas those who indicated that they were older than 40 years of age were assigned to the
older group.
With regard to the potential moderating effect of age on the relationship between
participation benefits and community participation, no moderating effects were confirmed.
H9a and H9d were tested, but not statistically supported. H9e (the path between monetary
benefit and participation) was not examined because the corresponding path was not
statistically significant in the causal model. H9f (the path between social-psychological
benefits and participation) was statistically significant, but was not supported due to the
opposite direction of the finding. Table 20 shows the moderating effect of age on the
relationship between each participation benefit and participation.
H9a posited that the effect of functional benefits on participation would be stronger
for the younger group than that for the older group. The path coefficients for the younger
members (p < .05) and older members (p < .01) were both significant, but the significance
level and the path coefficients were higher for the older members than those for the younger
members. However, the drop in χ2 after relaxing the restriction of equal path coefficients
across the two groups was .98, which did not exceed the minimum value of 3.84. Thus, this
hypothesis was rejected.
H9d posited that the effect of hedonic benefits on participation would be stronger for
younger members than for older members. The path coefficient between hedonic benefits
and participation was significant (p < .05) for older members and non-significant for
younger members. After relaxing the restriction of equal path coefficients across the two
groups, the drop in χ2 was 1.01, which did not exceed the minimum value of 3.84.
Therefore, this hypothesis was not statistically significant.
Since the path between social-psychological benefits and participation was a new
relationship identified after reducing measurement items, the present researcher posited H9f:
younger members of online communities are more likely than are older members to be
strongly affected by social-psychological benefits. The path coefficient between social-
psychological benefits and participation was significant for older members (p < .001), but
non-significant for younger members. The drop in χ2 after relaxing the restriction of equal

87
path coefficients across the two groups was 8.98, exceeding the minimum value of 3.84.
However, this finding is contrary to the proposed hypothesis, indicating that older members
were more strongly affected by social-psychological benefits. A finding opposite in the
direction to the proposed hypothesis leads to the rejection of that hypothesis (H9f).

88
Table 20. Moderating effects of age on the relationship between participation benefits
and participation in hotels’ Facebook pages
Path Unstandardized path
coefficients
Drop in
2 χ
p-
value
Young
(n=114)
Older
(n=79)
H9a Functional benefits→
participation
.39* .63** .98 -
H9d Hedonic benefits→ participation .18 .33* 1.01 -
H9e Monetary benefits→ participation - - -
H9f Social-psychological benefits→
participation
.17 .55*** 6.29* < .05
Note: * p < 0.05; ** p < 0.01; *** p < 0.001
With regard to the moderating effect of biological gender on the relationship
between participation benefits and participation, statistically significant differences in
biological gender were found in two paths: the relationship between functional benefits and
community participation (H10a) and between social-psychological benefits and community
participation (H10f). H10a was supported, whereas H10f was not supported due to the
opposite direction of the finding. H10d was not significant. H10e was not tested due to the
rejection of the previous causal relationship. Table 21 shows the moderating effects of
biological gender on the relationship between each participation benefit and participation.
H10a posited that the effect of functional benefits on participation would be stronger
for males than for females. The path coefficient between functional benefits and
participation was significant (p < .001) for males and non-significant for females. In
addition, the drop in χ2 after relaxing the restriction of equal path coefficients across the two
groups was 7.37, exceeding the minimum value of 3.84. This result indicated that males

89
seek out more functional benefits than do females. Therefore, this hypothesis was
supported.
H10d posited that the effect of hedonic benefits on community participation would
be stronger for male members. The path coefficient between hedonic benefits and
participation was significant (p < .05) for males and non-significant for females. However,
the drop in χ2 after relaxing the restriction of equal path coefficients across the two groups
did not exceed the minimum value of 3.84. Therefore, this hypothesis was not statistically
supported.
For the same reason indicated in H9f, the present researcher posited H10f: the effect
of social-psychological benefits on online community participation would be stronger for
females. The path coefficient between social-psychological benefits and participation was
significant (p < .01) for males and non-significant for females. The drop in χ2 after relaxing
the restriction of equal path coefficients across the two groups was 4.28, exceeding the
minimum value of 3.84. However, this finding is contrary to the proposed hypothesis,
indicating that male members were more strongly affected by social-psychological benefits.
Due to the opposite finding, H10f was rejected.
Table 21. Moderating effects of biological gender on the relationship between
participation benefits and participation in hotels’ Facebook pages
Path Unstandardized path
coefficients
Drop in
2 χ
p-
value
Male
(n=101)
Female
(n=94)
H10a Functional benefits→
participation
.89*** .27 7.37** < .01
H10d Hedonic benefits→ participation .32* .21 0.44 -
H10e Monetary benefits→ participation - - -

90
H10f Social-psychological benefits→
participation
.36** .11 4.28* < .05
Note: * p < 0.05; ** p <0.01; *** p < 0.001
Testing the structural model for the restaurant study
Which focused on the relationships among the proposed constructs. As performed through
SEM for the hotel study, social and psychological constructs were incorporated into a single
construct (social- psychological benefits). Accordingly, H2 and H3 were removed, and H11
was added as a new path, which illustrated the relationship between social-psychological
benefits and participation. In other words, it was proposed that social-psychological benefits
would have a positive influence on community participation.
Among the seven paths proposed in the conceptual model, five paths were
statistically significant, which included the path from the hedonic benefits to participation (β
= .178, t = 2.44, p < .05), from social-psychological benefits to participation (β = .27, t =
3.28, p < .05), from participation to brand commitment (β = .17, t = 2.23, p < .001), from
participation to brand trust (β = .20, t = 3.35, p < .001), and from brand trust to brand
commitment (β = .25, t = 3.17, p < .05). These results statistically supported H4, H6, H7,
H8, and H11. In other words, consumers‘ participation benefits (social-psychological and
hedonic benefits) positively influence consumer participation, which leads to consumer trust
and commitment toward a particular restaurant brand.
Two hypotheses were rejected: H1, predicting a positive effect of functional benefits
on consumer participation and H5, predicting a positive effect of monetary benefits on
consumer participation. the summary of this causal model is illustrated in Table 2

93
Table 22. Summary of support for hypotheses based on the results of SEM in the
conceptual model (restaurant study)
Hypothesis Path Proposed
effect
Result
H1 Functional benefits→ Participation + n.
H4 Hedonic benefits→ Participation + s.
H5 Monetary benefits→ Participation + n.
H6 Participation→ Brand commitment + s.
H7 Participation→ Brand trust + s.
H8 Brand trust→ Brand commitment + s.
H11 Social-Psychological benefits→ Participation + s.
Note: n. = non-significant; s. = significant
Testing the fully recursive model for the restaurant study
A fully recursive model including all plausible paths was constructed and estimated.
The model generated a total of 15 paths, with eight paths more than the original conceptual
model. The fully recursive model was significant (χ² = 337.03, df = 188, TLI = .95, CFI
= .96, RMSEA = .049). The chi-square ratio (χ2/df) was 1.79, which fell within McIver and
Carmines‘ (1981) acceptable range of 1 and 3. The fully recursive model decreased the χ²
values to 57.60 with 8 degrees of freedom, which was statistically significant at p < .001. In
comparison with the original conceptual model, the fully recursive model indicated a better
fit for the goodness-of-fit indicators. From the results, the fully recursive model appeared to
be more suitable than the conceptual model (Table 23).

94
Table 23. Chi-square test of model comparison for the restaurant study
Model comparison 2 χ df χ2/df TLI CFA RMSEA
Conceptual model 394.63 196 2.01 .94 .95 .055
Fully recursive model 337.03 188 1.79 .95 .96 .049
Δχ2(df) 57.60 (8)
p
< .001
By testing the fully recursive model, the present study identified four positive paths
and two negative paths that were statistically significant. Four paths were positive, including
the path from social-psychological benefits to participation (β = .26, t = 3.20, p < .001), the
path from hedonic benefits to participation (β = .18, t = 2.49, p < .05), the path from social-
psychological benefits to brand trust (β = .23, t = 2.71, p < .05), and the path from brand trust
to brand commitment (β = .29, t = 3.55, p < .001). However, two paths were negative,
including the path from hedonic benefits to brand commitment (β = -.18, t = -2.08, p < .05)
and the path from monetary benefits to brand commitment (β = -.38, t = -4.21, p < .001).
Among these, the path from social-psychological benefits to brand trust was newly identified
through the fully recursive model (H12b). In addition, the two negative paths from hedonic
benefits to brand commitment (H13) and from monetary benefits to brand commitment
(H14) were also identified as additional paths.
On the other hand, the paths that were statistically significant in the conceptual model
were not found to be significant in the fully recursive model. H6, which posited that online
community participation has a positive influence on brand commitment, was not supported
(p = .13). H7, which proposed that online community participation has a positive influence

95
on brand trust, was not statistically significant (p =. 17). Although these two hypotheses
turned out to be non-significant, the structural model with its newly identified paths indicated
significantly improved model fit indices. However, the present researcher accepted the
conceptual model because the model made better sense than the fully recursive model. In the
chapter 5, these new paths will be discussed with possible explanations because they are
important findings for future studies. Table 24 shows path coefficients and t-values for each
path in the reduced (theoretical) model and the fully recursive mode

97

Table 24. Unstandardized path coefficients and t-Value for structural model (restaurant
study)
Reduced (theoretical) model Fully recursive model
Community
participation
Brand trust Brand
commitment
Community
participation
Brand trust Brand
commitment
Functional
benefit
.08(.68)
b (t) b (t)
.08(.70) .01(.10) .06(.69)
Social-
psychological
benefit
.26(3.28) .21(3.20) .15(2.71) .02(.34)
Hedonic benefit .14(2.44) .15(2.49) .03(.62) -.10(-2.08)
Monetary benefit -.06(-1.05) -.06(-1.02) .07(1.40) -.26(-4.21)
Community
participation
.17(3.35) .15(2.23) .08(1.38) .08(1.51)
Brand trust .26(3.17) .23(3.55)
R2 .18 .04 .11 .18 .10 .27
Model fit χ² = 394.63, df = 196, TLI = .94, CFI = .95,
RMSEA = .055
χ² = 337.03, df = 188, TLI = .95, CFI = .96,
RMSEA = .049

98
Testing for mediating effects
The mediating role of brand trust on the relationship between community
participation and brand commitment was investigated with the analysis procedure of Barons
and Kenny‘s (1986). To test the mediating effect of brand trust, the structural equation
model was re-estimated by constraining the direct effect of brand trust on brand commitment
(the path coefficient was constrained to zero). The first condition would be met if
community participation (the independent variable) was found to influence brand trust (the
mediator variable), β21. The second condition would be satisfied if brand trust (the mediator
variable) affected brand commitment (the dependent variable), β32|1. The third condition
would be satisfied if community participation (the independent variable) influenced brand
commitment
(the dependent variable), β31|2. These three conditions were met in the original conceptual
model, given that all three paths were significant. The fourth condition would also be met if
the parameter estimate between community participation and brand commitment (β31|2
= .15*, t = 2.23) in the mediating model became less significant (partial mediation) than the
parameter estimate (β31 = .25***, t = 3.26) in the constrained model (Table 25). The results
showed that brand trust had a partial mediating role. In addition, the difference in the χ2
between the mediating model (χ2 = 394.63, df = 196) and the constrained model (χ2 =
404.77, df = 197) was statistically significant (χ2 = 10.14, df = 1). Thus, the mediating effect
of brand trust clearly demonstrates that members‘ community participation favorably affects
brand commitment through brand trust.
The indirect effect of community participation on brand commitment through brand
trust was .048 (β21 * β32|1 =.19 *.25). Even though the indirect effect was less than the direct
effect between participation and brand commitment (β31|2 = .18), this result emphasized the
role of brand trust as a mediating variable between community participation and brand
commitment.

99
Table 25. Mediating effects of brand trust in restaurants’ Facebook pages
Mediating model Constrained model
Participation Brand trust Standardized path coefficient
.19***
Participation Brand commitment .18* .22***
Brand trust Brand commitment .25** -
Indirect effect .048
Total effect .22
Note: * p < 0.05; ** p <0.01; *** p <0.001
Testing the moderating effects of age and biological gender for the restaurant study
In order to examine the impact of the moderator variable of age, a median age was
calculated to classify younger and older sub-groups. Based on the median age, participants
who indicated their age as under or equal to 30 years were assigned to the younger group,
whereas those indicating an age older than 30 years were assigned to the older group. In
order to evaluate the moderating effects of age and biological gender, the same procedures
utilized in the hotel study were conducted. Table 26 and Table 27 show no moderating
effects of either age or biological gender in the restaurant study. H9a and H9e were not
tested due to the insignificant findings in the causal model.

100
H9d posited that the effect of hedonic benefits on participation would be stronger for
the younger group than for the older group. The path coefficients between hedonic benefits
and participation were significant for both the groups at the same significant level (p < .05).
However, the drop in χ2 after relaxing the restriction of equal path coefficients across the two
groups did not exceed the minimum value of 3.84. Therefore, H9d was rejected.
To provide greater detail with regard to the effects of age, H9f posited that younger
members are more likely to be strongly affected by social-psychological benefits than are
older members. The path coefficients between social-psychological benefits and
participation were significant for both the younger group (p < .01) and for the older group (p
< .05).
However, the drop in χ2 after relaxing the restriction of equal path coefficients across the two
groups was .057, which did not exceed the minimum value of 3.84. Thus, this hypothesis
was not statistically significant.
Table 26. Moderating effects of age on the relationship between participation benefits
and participation in restaurants’ Facebook pages
Path Unstandardized path
coefficients
Drop in
2
χ
p-value
Younger
(n=210)
Older
(n=116)
H9a Functional benefits→
participation
- - -
H9d
Hedonic benefits→
participation .15* .22* .57 -
H9e
Monetary benefits→
participation
- - -

101
H9f Social-psychological
benefits→ participation .21** .19* .057 -
Note: * p < 0.05; ** p < 0.01
H10d posited that the effect of hedonic benefits on participation would be stronger
for males than for females. The path coefficient between hedonic benefits and participation
was significant (p < .05) for females and non-significant for males. The drop in χ2 after
relaxing the restriction of equal path coefficients across the two groups did not exceed the
minimum value of 3.84. Accordingly, this hypothesis was not statistically significant.
H10f posited that the effect of functional benefits on participation would be stronger
for males than for females. Table 27 shows that the path coefficients between social-
psychological benefits and participation were significant for both male (p < .05) and female
groups (p < .01). However, the drop in χ2 after releasing the restriction of equal path
coefficients across the two groups was .019, which did not exceed the minimum value of
3.84. Thus, this hypothesis was rejected.
Table 27. Moderating effects of biological gender on the relationship between
participation benefits and participation in restaurants’ Facebook pages
Path Unstandardized path
coefficients
Drop in
2 χ
p-
value
Male
(n=97)
Female
(n=230)
H10a Functional benefits→
participation
- - -
H10d Hedonic benefits→ participation .18 .14* .13 -
H10e Monetary benefits→ participation - - -

102
H10f Social-psychological benefits→
participation
.21* .20** .019 -
Note: * p < 0.05; ** p < 0.01
Summary
A brief summary of the research findings is provided. This study examined the
relationships between consumers‘ participation benefits derived from the visits to brands‘
Facebook pages and behavioral outcomes (community participation, brand trust, and brand
commitment). Responses to Facebook pages for hotels and for restaurants were examined
separately. Each study was conducted in two steps: 1) investigating the causal relationships
reflected in the hypotheses of the study, and 2) examining the moderating effects of age and
biological gender on the relationships between benefits and community participation. Table
28 shows the results regarding causal relationships and the results of moderating effects in
the hotel segment. Table 29 illustrates the results for the restaurant segment.

Table 28. Result of hypotheses tests for the hotel study
Hypothesis Path Proposed effect Result
Proposed model
H1 Functional benefits→ Participation + s.
H2 Social benefits→ Participation n.t.
H3 Psychological benefits→ Participation n.t.
H4 Hedonic benefits→ Participation + s.
H5 Monetary benefits→ Participation + n.s.
H6 Participation→ Brand commitment + n.s.
H7 Participation→ Brand trust + s.
H8 Brand trust→ Brand commitment + n.s.
H11 Social-Psychological benefits→ Participation + s.
Fully recursive model
H12a Monetary benefits→ Brand trust + s.
Note: n.t. = not tested; s. = significant; r. = significant, but in a reverse direction to the original hypothesis; n.s. = non-significant
Table 28. (continued)
103

Moderating variable: Age
H9a Functional benefits→ Participation n.s.
H9b Social benefits→ Participation n.t.
H9c Psychological benefits→ Participation n.t.
H9d Hedonic benefits → Participation n.s.
H9e Monetary benefits→ Participation n.t.
H9f Social-Psychological benefits→ Participation
Moderating variable: Biological gender
r.
H10a Functional benefits→ Participation s.
H10b Social benefits→ Participation n.t.
H10c Psychological benefits→ Participation n.t.
H10d Hedonic benefits→ Participation n.s.
H10e Monetary benefits→ Participation n.t.
H10f Social-Psychological benefits→ Participation r.
Table 29. Result of hypotheses tests for the restaurant study
Hypothesis Path Proposed effect Result
Proposed model
H1 Functional benefits→ Participation + n.s.
104 105

H2 Social benefits→ Participation n.t.
H3 Psychological benefits→ Participation n.t.
H4 Hedonic benefits→ Participation + s.
H5 Monetary benefits→ Participation + n.s.
H6 Participation→ Brand commitment + s.
H7 Participation→ Brand trust + s.
H8 Brand trust→ Brand commitment + s.
H11 Social-Psychological benefits→ Participation + s.
Fully recursive model
H12b Social-psychological benefits→ Brand trust + s.
H13 Hedonic benefits→ Brand commitment - s.
H14 Monetary benefits→ Brand commitment - s.
Note: n.t. = not tested; s. = significant; r. = significant, but in a reverse direction to the original hypothesis; n.s. = non-significant
Table 4.29 (continued)
Moderating variable: Age
H9a Functional benefits→ Participation n.t.
H9b Social benefits→ Participation n.t.
106

H9c Psychological benefits→ Participation n.t.
H9d Hedonic benefits→ Participation n.s.
H9e Monetary benefits→ Participation n.t.
H9f Social-Psychological benefits→ Participation
Moderating variable: Biological gender
n.s.
H10a Functional benefits→ Participation n.t.
H10b Social benefits→ Participation n.t.
H10c Psychological benefits→ Participation n.t.
H10d Hedonic benefits→ Participation n.s.
H10e Monetary benefits→ Participation n.t.
H10f Social-Psychological benefits→ Participation n.s.

107
CHAPTER 5. DISCUSSION AND CONCLUSIONS
This chapter discusses the interpretations of the findings. Conclusions, implications,
limitations, and recommendations for future research are also presented in this chapter.
Discussion of Findings
Relationships between participation benefits and community participation
In the hotel study, functional, hedonic, and social-psychological benefits (H1, H4, and
H11) from hotel brands‘ Facebook page members were found to positively influence
community participation. In contrast, the results from the restaurant study indicated that
community participation was significantly influenced by hedonic and social-psychological
benefits (H4 and H11), but not functional benefits (H1). In both studies, monetary benefits
were found to be a non-significant factor in community participation (H5).
The positive relationship between functional benefits and community participation for
hotel pages (H1) is consistent with the findings of Chung and Buhalis (2008) and Hwang and
Cho (2005), who indicated functional benefits as the most influential factors affecting the
level of members‘ participation in online travel communities. In relation to functional
benefits, members of hotel brands‘ Facebook pages in the present study desired efficiency
and convenience of communicating with others online, and desired sharing information about
their service experiences with the hotel brands. To fulfill these desires, members appeared to
visit the site frequently to gather information and communicate with others regarding the
hotel and its services. Accordingly, the findings of the present study identified functional
benefits as a significant element that increased member participation in the hotel brands‘
Facebook pages.
The non-significant relationship between functional benefits and community
participation for restaurant pages (H1) is consistent with the findings of Wang and
Fesenmaier (2004b), who reported that the functional benefits of online travel communities
were not a primary reason that members increase their visiting frequencies. One possible
explanation for this is that people may utilize other resources to obtain information about

108
restaurants, such as restaurant review sites and friends‘ referrals (O'Connor, 2009).
Therefore, the Facebook page may not be the only outlet from which to receive desired
functional benefits.
Hedonic benefits were found to be a significant motivating factor for community
participation in both the hotel and restaurant studies (H4). This supports previous findings
that indicate that members participate in community activities because they perceive these to
be relaxing and entertaining (Ridings & Gefen, 2004; Wasko & Faraj, 2000). Members are
likely to spend more time, especially when hotel or restaurant brands‘ pages incorporate
unique features that are geared toward members‘ interests and that give members another
way to interact (Dholakia et al., 2004).
Monetary benefit was a new construct added that extends Wang and Fesenmaier‘s
(2004b) conceptual model. Contrary to past research looking at book clubs and airlines
(Peterson, 1995), monetary benefits did not have a significant relationship with community
participation in either the hotel or restaurant study (H5). The present results also conflicted
with the results from Treadaway and Smith (2010) and Harris, O'Malley, and Patterson
(2003). Treadaway and Smith (2010) found that monetary benefits potentially help generate
member interest about hotel and restaurant brands and encourage members to participate in
community activities. Harris et al. (2003) reported monetary benefits as consumers‘ primary
reason to begin a relationship with a company. One possible reason for the conflicting results
is because the present study focused on taking part in activities rather than generating initial
interest or joining the community. For this reason, monetary benefits need to be more
thoroughly investigated to determine if they can stimulate potential consumers to join hotel
or restaurant brands‘ pages. In other words, monetary benefits can be an influential factor
that increases the number of members, but not necessarily the level of subsequent
participation.
According to the data analysis for H11, social-psychological benefits were composed
of two components, social benefits (getting involved with other members) and psychological
benefits (seeking a sense of affiliation and belonging in the community). This analysis
indicates that consumers do not make a distinction between social and psychological benefits;
rather, they perceive them to be a single benefit factor. In other words, members of the hotel
or restaurant brands‘ Facebook pages sought both psychological attachment to the

109
community and social relationships with other members. This merger of social and
psychological benefits aligns with past research (i.e., Chung & Buhalis, 2008; Lee, 2005).
The significant relationship between social-psychological benefits and community
participation in this present study also confirms the findings of previous studies (e.g., Ahuja
& Galvin, 2003; Langerak, Verhoef, Verlegh, & Valck, 2003). These social-psychological
benefits may also enhance the perceptions of community attractiveness and lead to useful
feedback about community service (Bendapudi & Berry, 1997).
Relationships between community participation, brand trust, and brand commitment
H6, H7, and H8 delineated the relationships between community participation, brand
trust, and brand commitment. The underlying assumption of these relationships posited that
active interaction with a particular brand evokes emotional attachment in its consumers and
enhances member trust toward the brand, which in turn, influences the development of a
deeper attachment to the brand (Thompson, MacIinnis, & Park, 2005). In the hotel study, the
relationship between community participation and brand trust (H7) was supported, whereas
the relationships between community participation and brand commitment (H6), and between
brand trust and brand commitment (H8), were not supported. In contrast, the results from the
restaurant study indicated that these three proposed relationships (H6, H7, and H8) were
supported.
The positive effect of community participation on brand trust (H7) was found in the
hotel study. The present study found that active participation in community activities (e.g.,
posting and reviewing hotel information and service experiences and actively participating in
community activities) was associated with trust toward the hotel brand. This result supports
the finding of Casalo et al. (2007), who reported that participation in community activities
fosters consumer trust. Specifically, Bagozzi and Dholakia (2006) and Ha and Perks (2005)
reported that consumers who are highly involved in community activities tend to build trust
toward the online community and the brand because consumers support each other‘s use of a
brand‘s product.
The positive effect of participation on brand commitment (H6) and the positive effect
of brand trust on brand commitment (H8) were not significant in the hotel study. These
results indicated that community participation and brand trust did not produce a positive

110
feeling of attachment to a brand. The results for H6 are contrary to the results of Jang et al.
(2008) and Casaló, Flavian, and Guinaliu (2010), who found a positive effect of member
participation on commitment toward a brand. The results for H8 also contradicted previous
relationship marketing literature that indicated that brand trust significantly influences strong
personal attachment and commitment of community members (Chaudhuri & Holbrook, 2001;
Ellonen, Tarkiainen, & Kuivalainen, 2010). The findings of the present study are noteworthy
in that neither participation nor brand trust had a significant impact on brand commitment
among hotel respondents. This aligns with the research on brand commitment (Matzler,
Grabner-Krauter, & Bidmon, 2006) that showed that promotions and promise of benefits
offered by competitors lured hotel consumers to switch to a competitor‘s brand.
One possible explanation for the behavior of hotel respondents is the amount of time
dedicated to interaction between members and brands. Ellonen et al. (2010) emphasized the
role of online consumer-brand interactions in strengthening consumer relationships with a
particular brand. They pointed out two key influential factors of consumer-brand
relationships: frequent participation and longer individual visits to online brand communities.
According to Merisavo and Raulas (2004), positive emotional responses occur as consumers
increase the amount of time that they spend with a brand, which enhances the relationship
with the brand (Merisavo, 2008). In contrast to these findings (Merisavo & Raulas, 2004;
Merisavo, 2008), member profiles of hotel respondents in the present study showed that over
70% of respondents had relatively short-term relationships with hotel brands‘ Facebook
pages (e.g., the duration of membership was less than a year) and nearly 90% spent less than
an hour per day participating in activities on hotel brands‘ pages. This indicates that members
of hotel brands‘ pages have relatively low levels of interaction with hotel brands. In addition,
the hotel study identified functional benefits as the most influential factor overall on
community participation (path coefficient = .31). This indicates that members may intend to
visit hotel brands‘ pages more often when they need to fulfill specific needs, such as
obtaining information about hotel packages and events. Because of the goal-oriented
behaviors of hotel members (pursuing specific needs for a special occasion such as a trip),
the members‘ visits to the hotel brands‘ pages are inclined to be infrequent rather than
consistent.

111
Unlike the hotel study, the results from the restaurant study supported the three
hypothesized relationships: the positive effect of participation on brand commitment (H6),
the positive effect of participation on brand trust (H7), and the positive effect of brand trust
on brand commitment (H8). These results support the findings of (a) Casaló et al. (2007),
who found that participation positively affected trust and commitment toward community
brands in the context of the online community of free software; (b) Holland and Baker
(2001), who revealed a significant relationship between frequency of visits to brand sites and
brand loyalty in the context of corporate websites; and (c) Ha (2004), who confirmed the
positive effect of brand trust on brand commitment in the online business context. Overall,
the results of the restaurant study indicate that participation in restaurant brands‘ pages may
evoke positive emotional responses in the minds of members and strengthen their trust in
restaurant brands, which, in turn, helps build a strong relationship between members and
restaurant brands.
With regard to the mediating effects of brand trust (Table 4.18), the indirect effect of
community participation on brand commitment (path coefficient = .048) was weaker than the
direct effect (path coefficient = .18). Although brand trust did not strengthen the effect of
participation on brand commitment, this finding reveals that community participation not
only directly influenced brand commitment, but also indirectly influenced brand commitment
through brand trust.
Moderating effects of age and biological gender
H9 and H10 postulated the moderating effects of age and biological gender on the
relationships between participation benefits and community participation. Based on the
results of the causal model, H9a and H10a were not tested in the restaurant study, because the
paths between functional benefits and community participation were not significant. In
addition, H9e and H10e were not tested in either the hotel or the restaurant studies for the
same reason.
With regard to the moderating effect of age, the findings of the present study did not
provide evidence to support the effect of age on the relationship between participation
benefits and community participation. Specifically, the hotel study rejected the moderating

112
effect of age on the relationship between functional benefits and community participation
(H9a). The proposed moderating effects of age on the relationship between hedonic benefits
(H9d) or social-psychological benefits (H9f) and community participation were rejected in
the hotel and restaurant studies, respectively. Interestingly, H9f was rejected because the
effect of age was significant but in the opposite direction from what was hypothesized.
Similar to prior studies (White, 2008; Zaphiris & Rifaht, 2006), the impact of social-
psychological benefits on community participation was stronger for older members than
younger members.
With respect to the moderating effect of biological gender, the impact of functional
benefits on community participation was stronger for males than females in the hotel study
(H10a). This same relationship was not tested in the restaurant study. The finding of the hotel
study supports several previous studies (Mo, Malik, & Coulson, 2009; Phillip & Suri, 2004)
that found males to be more likely than females to participate in information search and
practical tasks online. The moderating effect of biological gender on the relationships
between hedonic benefits (H10d) or social-psychological benefits (H10f) and community
participation were not supported in either the hotel or restaurant study. H10f of the hotel
study, which posited that the effect of social-psychological benefits on community
participation would be stronger for females than for males, was rejected due to a significant
path in the opposite direction of what was proposed. The result of the hotel study indicated
community participation was more strongly affected by social-psychological benefits for
males than for females. These findings illustrating few moderating effects of age and
biological gender reinforce previous studies (Jones & Fox, 2009; Ono & Zavodny, 2005) that
revealed that the differences in online behavior caused by age and biological gender have
disappeared over time or have reversed.
The non-significant moderating effects of age are consistent with previous studies
(Hernández, Jiménez, & Martin, 2011; Jones & Fox, 2009) that found that age is not an
obstacle that prevents people from using the Internet. With regard to participation in online
communities, older people tended to seek information, make purchases, and build social
networks (Jaeger & Xie, 2009; Jayson, 2009). From this perspective, older members may
seek hotel information, share experiences with others, and feel affiliation with the group
through participating in the activities on Facebook pages. Accordingly, the present research

113
argues that the community benefits perceived by older members are similar to those
perceived by younger members.
Additional paths between participation benefits, brand trust, and brand commitment
A fully recursive model was tested with all the possible paths in the conceptual model.
The results generated several unexpected paths that were statistically significant. Specifically,
the hotel study identified a direct path from monetary benefits to brand trust (H12a). Because
the inclusion of this path significantly improved the model fit of TLI, CFA, and RMSEA (see
Table 4.13), which was low in the finalized conceptual model, this new path is considered
important when examining the impact of member benefits on brand trust.
Prior studies showed that consumers feel monetary loss and/or insecurity about
products or services purchased online when the products/services do not meet their
expectations, thus resulting in a decrease in consumer brand trust (Alam & Yasin, 2010; Ha,
2004). Because brand trust refers to consumers‘ secure belief in a brand‘s ability to perform
as promised, it can be achieved when a brand satisfactorily fulfills consumer needs (Ellonen
et al., 2010; Ha & Perks, 2005). With the new path identified between monetary benefits and
brand trust, the present study suggests that members of hotel brands‘ Facebook pages may
experience an increase in trust in the brand only when they receive the services or products
promised in special offers.
The restaurant study identified three unexpected paths: a positive significant path
from social-psychological benefits to brand trust (H12b); and two significant negative paths,
from hedonic benefits to brand commitment (H13), and from monetary benefits to brand
commitment (H14). Despite these newly identified paths, the present researcher suggests that
the conceptual model is a better model, because the model fit indices of the fully recursive
model were only slightly improved in comparison to those of the finalized conceptual model
(see Table 23), but this slight improvement came at the cost of losing two significant
hypothesized paths of the conceptual model (H6: community participation -> brand trust and
H7: community participation -> brand commitment). In addition, the fully recursive model
identified two negative paths that may require further analysis in order to arrive at
satisfactory explanations.
H12b, which posited that social-psychological benefits have a positive influence on
brand trust, was newly identified. This finding aligns with previous research. For instance,

114
Bove and Johnson (2000) revealed that highly affiliated members are likely to show high
levels of trust in service employees. Social-psychological benefits are achieved through
communication and shared experiences (Wang & Fesenmaier, 2004b). Information shared by
other members affects a member‘s level of trust in the community because of high credibility
given to trusted members (Stockdale & Borovicka, 2006; Watson, Morgan, & Hemmington,
2008). According to Stockdale and Borovicka (2006), the members of online communities
feel a strong sense of social belonging when viewing postings (messages or contents by
others in the community), if the content reveals similar interests and views. This similarity of
view can increase the sense of trust (Ridings et al., 2002).
For H13, a negative relationship was found between hedonic benefits and brand
commitment toward a restaurant. This indicates that the importance of hedonic benefits, such
as feelings of entertainment and enjoyment while engaging in community activities,
negatively influenced consumers‘ brand commitment. One possible explanation may be
found in McAlister and Pessemier‘s study (1982); the desire for hedonic benefits may
encourage consumers to seek variety in the given product category. Similar behavior has
been found for hedonic experiences (Dodd, Pinkleton, & Gustafson, 1996). Consumers may
switch to other brands to obtain new hedonic experiences of fun and excitement from
products or services (Chandon, Wansink, & Laurent, 2000). Thus, restaurant brands that offer
a higher level of hedonic benefits may be less likely to maintain loyal consumers (Carroll &
Ahuvia, 2006) because these consumers may seek hedonic experiences through visiting a
variety of restaurant Facebook fan pages. Variety-seeking behavior, associated with hedonic
pleasure, may also lead members to switch to Facebook pages of other brands. The present
study supports this argument because 86.6% of restaurant respondents joined two or more
restaurant brands‘ Facebook pages (see Table 4.2).
A negative relationship between the importance of monetary benefits and brand
commitment was found for H14. In other words, the higher the importance of monetary
benefits, the lower the brand commitment. Because monetary benefits in the present study
were defined as monetary savings such as discounts or special price breaks, it is not
surprising to learn that members of a restaurant brand Facebook page are less likely to feel
committed toward the brand when monetary benefits could be found on Facebook pages of
numerous brands.

115
Conclusions and Implications
The purposes of this present study were to (a) identify online hotel and restaurant
Facebook page members‘ participation benefits; (b) examine the relationships between
members‘ levels of participation, brand trust, and brand commitment; and (c) investigate the
moderating effects of demographic characteristics (i.e., age and biological gender) on the
relationship between participation benefits and community participation. First, the results
produced different sets of community benefits for the hotel and restaurant segments. These
findings suggest that the marketers in the two segments need to use different approaches to
manage their brand pages in social media. Second, the outcomes of community participation
were also different between the two segments, which emphasizes that there were differences
in consumer behavior associated with hotel and restaurant brands. Hospitality marketers for
hotels and restaurants may need to apply different marketing strategies to build brand
relationships with their respective consumers.
The study of hotel or restaurant brand use of social media, particularly in the context
of Facebook pages, is relatively new in the area of hospitality marketing. The present study is
the first to empirically examine benefits from member participation in brand communities
managed by hospitality firms and to investigate the impact of this participation on consumer
responses. These responses have marketing implications for each hospitality segment,
including the design of hotel or restaurant brand Facebook pages and other variables (e.g.,
brand awareness, brand loyalty, and perceived quality) for practitioners to attract potential
consumers and strengthen relationships with current consumers.
This study also examined the outcome variables of community participation (brand
trust and brand commitment), which strengthen consumer relationships with a particular
brand. Previous studies had indicated that community members were often persuaded by
other consumers to purchase and bond with a brand, which in turn built brand commitment
(Casaló et al., 2007; Ellonen et al., 2010). The results from the hotel study did not provide
evidence for the connections between community participation, brand trust, and brand
commitment. In contrast, the restaurant study confirmed a significant impact of community
participation on brand trust and brand commitment. These results of the present study provide
brand page marketers with insights into relationship marketing endeavors. Marketers of hotel

116
brands‘ pages must focus on a way to create relationships between consumers and the hotel
brand, whereas restaurant brands‘ page marketers need to consider a way to strengthen their
relationships with current members.
The present study proposed demographic characteristics of community members (age
and gender) as a significant potential factor in influencing member participation. The results
of the present study indicated that the strength of relationships between participation benefits
and community participation did not vary with age and biological gender. The role of
demographic characteristics in influencing consumer behavior seems to be nullified due to
increased user experience with social media. Therefore, the present research suggests that
marketers need to identify other factors that influence members‘ participation behavior for
their market segmentation.
Managerial implications for the hotel study
The present study suggests significant strategies for online community design by
identifying benefit factors that encourage member participation on hotel brands‘ Facebook
pages. The findings indicated three desired participation benefits for hotel members:
functional, social-psychological, and hedonic benefits. These benefits are related to multiple
consumer needs. From a managerial perspective, marketers of hotel brands‘ pages first need
to be aware of their members‘ characteristics and understand who their members are before

117
developing strategies for successful Facebook pages. In addition, these benefits can be used
to attract potential consumers to join the hotel brands‘ pages.
Hotel firms should provide communication devices with diverse formats (e.g., real-
time synchronous or asynchronous communication technologies such as chat or bulletin
boards, virtual product presentations) that enable members to exchange information about
hotel properties/services, provide critiques of ambiance, and share service experiences.
Information gathering through brand pages is the most influential element to attract potential
consumers to join hotel brands‘ Facebook pages and to encourage current members to
frequently visit the page.
Valuable information for the firm can be collected from the communication among
consumers. By analyzing the information, hotels may gain new insights into consumer trends,
needs, and experiences that affect (dis)satisfaction (Harwood & Gary, 2009). Hotel firms
should consider data-mining software to monitor the content of information posted by their
members. This would enable marketers to analyze the success of current marketing activities
and create opportunities to refine strategies, which in turn enhance business performance
(Kasavana, 2008; Fisher, 2011a). Through monitoring, marketers may provide ongoing
updates to brand-related information in order to satisfy members‘ current needs. Satisfying
members‘ information needs is important, because efficient and convenient information
gathering was a primary purpose of hotel brands‘ page members in the present study.
Hotels should enhance opportunities that help members identify like-minded
consumers who are seeking similar hotel services (e.g., in-room hotel technologies such as
touch screen tablets or Wi-Fi). It is the nature of an online community that individuals gather
together based on similar interests and purposes (Wang & Fesenmiar, 2004a). On a hotel
brand‘s page, individuals may form a variety of sub-groups based on similar or specific needs
for hotel services. Marketers need to identify these potential sub-groups and provide more
specialized and personalized services to each group (Kasavana, 2008).
In addition to categorizing sub-groups, marketers need to incorporate a variety of
tools in order to facilitate the hedonic nature of their brand pages. For example, hotels may
use a gaming platform (e.g., simple poll, online flash, online puzzles) for notifications of new
services. Adding videos related to new brand information and virtual tour devices gives

118
members enjoyable experiences during their visit to the hotel brands‘ pages (van Dolen &
Ruyter, 2002). New technologies including RFID (Radio-frequency identification) can be
employed to allow members to carry out community activities (e.g., photos taken at a hotel
are automatically posted on its brand Facebook page and tagged on members‘ own pages)
without the presence of a computer or smart-phone during their stay (Harbison, 2011). With
these features, hotel brands can enhance member engagement by increasing the hedonic
experiences of being a member of the hotel brands‘ pages and directly influence the positive
impression of brands. Therefore, it is critical that marketers implement various features that
enable members to enjoy all of the content on the brands‘ pages.
Hotels may also launch marketing campaigns that increase member participation in
their brands‘ pages by encouraging members to post messages and photos. At the same time,
they may reward the members for their participation with an element to enhance their
hedonic experience such as a free drink at check-in. These types of campaigns may be more
effective at engaging members in the activities of the brand page because they foster a
hedonic benefit such as enjoyable experiences, pleasure, and positive emotions. Moreover,
marketing campaigns with free gifts or free samples (e.g., Westin providing a signature-
scented candle) might be more effective over simple discounts and coupons in generating
favorable evaluation of a given brand (Chandon et al., 2000).
Interactions and communication among consumers do not appear to help hotel brands
develop consumer commitment. Marketers may need to devise methods that depend on
business-to-consumer activities rather than facilitating interaction among consumers in order
to build consumer commitment towards the brand. For example, direct communication via
online chat features between consumer service and consumers when making reservations may
be an effective approach to build consumer commitment. Through this process, hotel staff
could directly identify consumer preferences (e.g., the type of pillow or room they prefer) and
special requests (e.g., particular room temperature and particular newspaper they want to be
delivered) (Weed, 2011). Based on the information collected through this chat feature, hotels
may provide personalized service that underlines the value of staying at the hotel brand. This
may enhance consumer commitment toward a hotel brand and produce loyal customers.
Finally, social media is an innovative tool by which hotel brands can take a proactive
approach to manage brand relationships with their consumers. Hotel marketers can identify

119
the most important attributes of their brands (e.g., rooms, front desk, breakfast, room rates,
and cleanliness) by analyzing the consumers‘ posts on the brand pages. Following this
analysis, marketers can take immediate actions based on both complimentary reviews and
complaints about service. Particularly for uncomplimentary reviews, marketers can mitigate
their potential harmful effects by effectively responding to consumers‘ comments. All of
these efforts can assist hotel brands in creating favorable brand images and building strong
relationships with their members.
Managerial implications for the restaurant study
As in the hotel study, the findings of the restaurant study provided a number of
critical managerial implications to increase the success of restaurant brands‘ Facebook pages.
Two benefit factors, social-psychological benefits and hedonic benefits, were derived from
the restaurant consumers‘ need for participation. Restaurant members perceive a sense of
community and affiliation as the highest benefits they receive from restaurant brands‘
Facebook pages. In addition, they enjoy visiting restaurant brands‘ pages for fun and
entertainment. These benefits are important in helping hospitality firms to encourage the
involvement of the members in community activities that in turn may enhance brand
relationships with these current and potential consumers.
Marketers are advised to monitor the communication among the members on
restaurant brands‘ Facebook pages to gain insight. Restaurant attributes, such as the quality
and taste of food and employee service are frequently evaluated (Dellarocas, 2001). Active
participants are likely to post their personal thoughts and emotions regarding their dining
experiences; the content of these posts may influence other members (Green, 2009).
Marketers may identify groups of consumers categorized by certain criteria such as postings
about the taste of particular food items, preferences for food presentation, and positive or
negative opinions about new menu items. This may aid marketers in identifying the special
interests of their members and finding additional niche segments in existing markets.
Restaurants can develop new menus or items based on such content in order to meet the
specific needs of each target group.

120
Marketers of restaurant brands must provide numerous Internet-based opportunities
for members to share their experiences and interact with others (Watson, Morgan, &
Hemmington, 2008). For example, a restaurant may consider providing a personal space on
its Facebook page for active participants to post their own dining experiences, to which other
members could provide feedback. Because the postings and reviews from a personal page are
perceived as coming from trusted members of the community, other members would consider
them to be more credible than other review sites (Watson et al., 2008). By doing so, members
may experience enhanced positive feelings about the community of like-minded people and
may experience greater hedonic benefits of participation (Stockdale & Borovicka, 2006).
Moreover, marketers are advised to consider using a variety of tools such as games,
videos, and applications in order to create opportunities for positive experiences when
members visit. When using such entertainment tools, consumer engagement comes from
seeking hedonic experiences, a primary desire for community members. The purpose of
member visits is simply to play games or watch video, as opposed to participating in
community activities (Fisher, 2011b). Accordingly, games or videos should be developed
with the concept of entertaining members and sharing information about restaurants while
members are playing or viewing.
Although restaurant consumers appear not to be primarily driven to seek restaurant
information through Facebook pages, it is critical for restaurants to promote their menus and
entice potential consumers to visit their restaurants through this platform. To entice
consumers, marketers of restaurant brands‘ Facebook pages may consider providing special
offers and interesting content through the elements desired by Facebook users, such as
newsfeeds and widgets (Fisher, 2011b), which include updated information and make unique
impressions. .
As discussed previously, encouraging member participation on Facebook pages by
offering monetary benefits such as coupons and reduced prices to existing consumers appears
to be an ineffective approach. As indicated in previous studies (Buil et al., 2011; Chandon et
al., 2000), it is preferable to develop non-monetary promotions such as free beverages or free
sample menu items as compensation to existing members to encourage active participation.
However, monetary promotions can be used to entice potential consumers who are not yet
members of a brand page. Furthermore, a promotion should involve some activities to engage

121
consumers (e.g., announcing the winners of photo contests to provide social recognition
within the brand page). Facilitating monetary and non-monetary promotions with pleasurable
experiences can increase the popularity and success of a restaurant brand‘s Facebook page.
In conclusion, social media can be an effective platform for consumer engagement in
the restaurant industry. Restaurant brand pages enable both members and marketers to carry
out interactive communication. Marketers must design various marketing strategies that
strengthen consumer loyalty.
Summary
Social media provide a technology infrastructure that hospitality firms can embrace
with suitable planning and guidelines for consumer engagement. The present study found
several important benefit factors (i.e., social-psychological and hedonic benefits) that
influence member participation in both hotel and restaurant brands‘ pages. Marketers are
advised to provide these benefits to members of hotel or restaurant brands‘ Facebook pages.
Such strategies include:
• Enhancing opportunities for interaction and engagement among like-minded members
of a brand‘s Facebook page to foster sharing of interests and carrying out similar
purposes for joining.
• Incorporating various features into a brand‘s Facebook page that provide positive
experiences (e.g., entertainment, pleasure, and enjoyment) with the brand.
• Monitoring members‘ communications to identify new market segments and provide
customized services based on common interests about products/services, visiting
purposes, and other factors.
Consumer-generated content posted prior to, during, and following experiences with
hotel and restaurant brands is one of the most important resources that affect favorable brand
image and experiences. Hospitality firms may be tempted to keep only positive messages and
compliments on their pages and to delete negative content in order to create favorable brand
images and experiences. However, marketers should realize that brand pages that contain
only positive content are typically perceived with skepticism by visitors (Kasavana, Nusair,
& Teodosic, 2010). Brand page marketers can take advantage of negative comments by

122
giving satisfactory explanations on their Facebook pages and/or by following up with the
customer to rectify the problem, which may lead to the regaining of consumer trust. Such
efforts to address these issues can directly affect the consumers who post negative contents
and indirectly influence the others who view the communication.
Overall, a primary goal of brand page marketers using Facebook pages is to convert
the existing and new members into committed members in order to form long-term
relationships. Hospitality firms can foster brand commitment by providing specialized and
personalized services on Facebook (e.g., American express cards partners with several hotels
and restaurants to provide special deals based on member interests and likes in both
categories). In addition, hotels and restaurants need to be selective in order to provide the
right combination of services to an identified consumer segment. For example, a price
reduction offer may be attractive to new community members or consumers who are price
sensitive, but not effective for committed members who were found to be less price sensitive.
Companies can also treat committed members as a special group of consumers who receive
advance notice of new products or services, giving them exclusive opportunities for pre-
experience or pre-purchase comparisons (e.g., restaurants may invite their loyal consumers to
a sampling party) before other consumers. Since committed consumers are willing to spend
more, visit more frequently, and spread referral information to their friends, hospitality firms
may ultimately expect profit increases by successfully operating brand pages (Kasavana et
al., 2010).
Limitations and Future Study
The present study contains several limitations that should be identified and that lead
to suggestions for future research. First, the sample for the study was conducted from two
sources: an online research company and. The majority of hotel data were obtained from the
online research company, and the respondents had a range of education levels. Conversely,
the data for the restaurant study were obtained exclusively from the University, and the
majority of respondents were highly educated, holding either bachelor or graduate degrees.
These two sets of data were used in distinct studies; it is possible that the differences in the
results were caused by the demographic differences in the two groups of respondents. Future
research comparing hotel and restaurant Facebook pages may use one source of data and/or
ensure similar percentages of respondents from each demographic category are represented.

123
Scant empirical literature on online hospitality brand communities provides a rudimentary
foundation for the present study. Future hospitality research is needed to strengthen the
theoretical and empirical background that explicates the role of benefits from the firm‘s
Facebook pages on consumer responses.
Second, the measurement scales for participation benefits were highly correlated and
thus, multi-collinearity problems occurred during confirmatory factor analysis and SEM‘s
structural analysis. Six of the eighteen measurement items were eliminated to resolve this
issue. Future studies may focus on refinement and validation of the scales employed in the
present study to help marketers gain significant insight into the beneficial aspects of social
media communities valued by their target markets. In addition, the present study suggests the
need to improve measurement scales for brand commitment. In the hotel study, the factor
loading of one measurement scale, brand commitment, was larger than 1. This may be caused
by various reasons, which needs further analysis in future studies. In the restaurant study, the
Cronbach‘s alpha estimate for brand commitment (.60) was relatively low as compared to the
cutoff value of .70 (Nunnally, 1978). Future research needs to develop an internally
consistent measurement scale for brand commitment in the restaurant context.
Third, the present study identified two negative paths between hedonic benefits and
brand commitment and between monetary benefits and brand commitment. The two negative
paths may be caused by the multi-collinearity issues mentioned above. These results, in the
direction opposite from predicted, suggest future studies are needed to examine if negative
relationships truly exist, and if so, why hedonic and monetary benefits offered by a brand‘s
Facebook page negatively affect brand commitment. Future studies may conduct interviews
to determine consumer perceptions about hedonic and monetary benefits of online brand
communities and the direction of impact these benefits have on brand commitment.
Fourth, the present study found that age and biological gender did not play a
moderating role in evaluating the effects of benefits on member participation. Future studies
may employ other factors, such as involvement or personality traits, which may influence the
relationships in the conceptual model. For example, involvement with a brand may affect the
strength of the relationship between participation benefits and member participation (Lee,
2005; Tsao & Chang, 2010). Personality traits may also be considered as an influential
moderating variable, because certain personality traits can influence the relationship between

124
participation benefits and member participation on brand pages (Morse, 2009). Conducting
research using other moderating variables could help explain consumer participation behavior
in online brand communities.
Fifth, the path from monetary benefits to brand commitment in the hotel study should
be investigated further. The path direction may run in reverse between the two constructs,
with brand commitment affecting desired benefits rather than benefits affecting commitment.
Committed members may desire monetary benefits as a reward for their commitment toward
the hotel brand. Monetary benefits given to committed members may encourage them to
participate in various community activities, such as posting positive service experiences,
promoting brand pages to potential consumers, and supporting other members‘ opinions.
Thus, future research may consider the effect of brand commitment on benefits, which in turn
enhance member participation on brand pages.
Finally, the present study proposed a single final consequence of community
participation, which was brand commitment. Future research may investigate other
consequences of active participation such as brand loyalty, purchase intention, or brand
equity. This may provide hospitality companies with specific information needed to
implement marketing strategies that encourage more active member participation.

125
APPENDIX A: A QUESTIONNAIRE FOR THE HOTEL SEMENT
Screening Question: This study aims to investigate the experiences of fans on Facebook
pages of hospitality companies. Therefore, if you have no experience visiting the
Facebook pages of any hotels please quit the survey now. Otherwise, please choose
which Facebook pages you are a fan of. Hotel Restaurant Quit Survey
Please select only one of the following: Marriott Napa Valley Hotel and Spa Stanley Hotel Beacon Hotel The Algonquin Hotel Others. Specify
Please recall your prior experience on the specific hotel Facebook page. Then answer
the questions in the following sections.
Section 1: In this section, we are interested in the benefit you derive as a fan of a Hotel
brand‘s Facebook page. Please use the following scale to rate your level of agreement with
each statement.

126
Seeking enjoyment on this Facebook page
Strongly
agree
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Agree
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Neither
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Disagree
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Strongly
disagree
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I think the following is important…..
Obtaining up-to-date information about
the Hotel brand
Conveniently communicating with others
online
Efficiently communicating online
Sharing experiences in the Hotel brand
Having trust in the community of
Facebook
Establishing and maintaining relationships
with other members of Facebook
Communicating with other members
Getting involved with other members
Seeking a sense of affiliation in the
community
Seeking a sense of belonging
Seeking self-identity
Being amused by other members
Having fun on the brand‘s Facebook page
Being entertained on this Facebook page
Obtaining discounts or special deals that
most consumers don't get
Obtaining better prices than other

133
consumers
Receiving free coupons for the Hotel
brand by becoming a member of the
Facebook page
1 2 3 4 5

128
Section 2: We are interested in how you participate as a fan of the brand Hotel‘s Facebook
page and your thoughts about the brand. Please indicate how you agree or disagree with
each statement, using the following scale:
I take an active part in the brand
Hotel‘s Facebook page
I frequently provide useful information
to other members
Strongly
disagree
Disagree
2
2
Neither
3
3
Agree
4
4
Strongl
y agree
5
5
In general, I post
messages and
responses on the
brand‘s Facebook
page with great
enthusiasm and
frequency
1 2 3 4 5
I do my best to participate in activities
offered on the brand‘s Facebook page
What the Hotel brand says about its
products/service is true
I feel I know what to expect from the
Hotel brand
The Hotel brand is very reliable
The Hotel brand meets its promises
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
Section 3: We are interested in further thoughts about the Hotel brand you chose. Please
indicate the level of agreement with each statement using the following scale:
If the Hotel brand had no available
Strongly
disagree
DisagreeNeitherAgreeStrongl
y agree

129
reservations, I would have no problem2 3 4 5
finding a different Hotel with which I
would want to make reservations
I consider myself to be highly loyal to
the Hotel brand
When another brand has a special deal
(e.g., discounted room rate), I generally
visit that Hotel with the better deal
2
2
3
3
4
4
5
5

130
Section 4: Please indicate your level of agreement with each of the following statements.
There is active participation between the
company and members
The Hotel brand‘s Facebook page is
successful
I like visiting the Hotel brand‘s Facebook
page
Strongly
disagree
1
1
1
Disagree
2
2
2
Neither
3
3
3
Agree
4
4
4
Strongly
agree
5
5
5
Section 5: Demographic Information.
What is your biological gender?  Male  Female
How old are you?  18 - 20 years old  21 - 30  31 - 40  41 - 55  over 55
What is the highest level of education you have completed? High school or less  Associate degree  Bachelor‘s degree Graduate degree (Master‘s, J.D., M.D., or Doctoral)  Other degree
In what region of the world you do reside? Africa  Asia  Oceania  Europe United States  Canada  Central America  South America Middle East  Others. Specify
(1) How long have you been a member of this Hotel brand’s Facebook page?
months
(2) How long, on average, do you participate in this Hotel brand’s page each week?

131
hours
(3) How many Hotel Facebook pages are you a member of?
Please provide your email address if you want to be considered for drawings:
** Your email information will not be connected with your response.
Thank You Very Much

132
APPENDIX B: A QUESTIONNAIRE FOR THE RESTAURANT SEMENT
Screening Question: This study aims to investigate the experiences of fans on Facebook
pages of hospitality companies. Therefore, if you have no experience visiting the
Facebook pages of any hotels/ restaurants please quit the survey now. Otherwise,
please choose which Facebook pages you are a fan of. Hotel  Restaurant  Quit Survey

133
Please select only one of the following: Chili‘s Outback Red Lobster The Cheesecake Factory Others. Specify
Please recall your prior experience on the specific restaurant Facebook page. Then
answer the questions in the following sections.
Section 1: In this section, we are interested in the benefit you derive as a fan of a Restaurant
brand‘s Facebook page. Please use the following scale to rate your level of agreement with
each statement.
I think the following is important…..
Obtaining up-to-date information about
the Restaurant brand
Conveniently communicating with others
online
Efficiently communicating online
Sharing experiences in the Restaurant
brand
Having trust in the community of
Facebook
Establishing and maintaining relationships
with other members of Facebook
Communicating with other members
Getting involved with other members
Strongly
disagree
1
1
1
1
1
1
1
1
Disagree
2
2
2
2
2
2
2
2
Neither
3
3
3
3
3
3
3
3
Agree
4
4
4
4
4
4
4
4
Strongly
agree
5
5
5
5
5
5
5
5

134
Seeking a sense of affiliation in the
community
Seeking a sense of belonging
Seeking self-identity
Being amused by other members
Having fun on the brand‘s Facebook page
1
1
1
1
1
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5

135
5
4
4
4
4
4
3
3
3
3
3
2
2
2
2
2
1
1
1

136
1
1
5
Being entertained on this Facebook page
Obtaining discounts or special deals that
most consumers don't get
Obtaining better prices than other
consumers
Receiving free coupons for the Restaurant
brand by becoming a member of the
Facebook page
5
5
5
Seeking enjoyment on this Facebook page

137
Section 2: We are interested in how you participate as a fan of the x brand Restaurant‘s
Facebook page and your thoughts about the brand. Please indicate how you agree or
disagree with each statement, using the following scale:
I take an active part in the x brand
Restaurant‘s Facebook page
I frequently provide useful
information to other members
Strongly
disagree
1
1
Disagree
2
2
Neither
3
3
Agree
4
4
Strongly
agree
5
5
In general, I post
messages and
responses on the
brand‘s Facebook
page with great
enthusiasm and
frequency
1 2 3 4 5
I do my best to participate in activities
offered on the brand‘s Facebook page
What the Restaurant brand says about
its products/service is true
I feel I know what to expect from the
Restaurant brand
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5

142
The Restaurant brand is very reliable
The Restaurant brand meets its
promises
1
1
2
2
3
3
4
4
5
5

139
Section 3: We are interested in further thoughts about the Restaurant brand you chose.
Please indicate the level of agreement with each statement using the following scale:
Strongly
disagree
Disagree Neither Agree Strongly
agree
If the Restaurant brand had no available
reservations, I would have no problem
finding a different restaurant with
which I would want to make
reservations
1 2 3 4 5
I consider myself to be highly loyal to
the Restaurant brand
1 2 3 4 5
When another brand has a special deal
(e.g., discount price for meal), I
generally visit that restaurant with the
better deal
1 2 3 4 5
Section 4: Please indicate your level of agreement with each of the following statements.
There is active participation between the
company and members
The Restaurant brand‘s Facebook page is
successful
I like visiting the Restaurant brand‘s
Facebook page
Strongly
disagree
1
1
1
Disagree
2
2
2
Neither
3
3
3
Agree
4
4
4
Strongly
agree
5
5
5
Section 5: Demographic Information.
What is your biological gender?  Male  Female
How old are you?

140 18 - 20 years old  21 - 30  31 - 40  41 - 55  over 55
What is the highest level of education you have completed? High school or less  Associate degree  Bachelor‘s degree Graduate degree (Master‘s, J.D., M.D., or Doctoral)  Other degree

141
In what region of the world you do reside? Africa  Asia  Oceania  Europe United States  Canada  Central America  South America Middle East  Others. Specify
(1) How long have you been a member of this Restaurant brand’s Facebook page?
months
(2) How long, on average, do you participate in this Restaurant brand’s page each
week? hours
(3) How many restaurant Facebook pages are you a member of?
Please provide your email address if you want to be considered for drawings:
** Your email information will not be connected with your response.
Thank You Very Much

142

143
REFERENCES
Ahluwalia, R. (2000). Examination of psychological processes underlying resistance to
persuasion. Journal of Consumer Research, 27, 217–232.
Ahluwalia, R., Burnkrant, R. E., & Unnava, H. R. (2000). Consumer response to negative
publicity: The moderating role of commitment. Journal of Marketing Research, 37,
203–214.
Ahuja, M., & Galvin, J. (2003). Socialization in virtual groups. Journal of Management, 29,
161–185.
Ä kkinen, M., & Tuunainen, V. K. (2005). Conceptual foundations of online communities.
Retrieved from http://sprouts.aisnet.org/5-27
Alam, S. S., & Yasin, N. M. (2010). An investigation into the antecedents of customer
satisfaction of online shopping. Journal of Marketing Development and
Competitiveness, 5(1), 71-78.
Antikainen, M. (2007). The attraction of company online communities. A multiple case study.
Retrieved from http://acta.uta.fi/pdf/978-951-44-6850-6.pdf
Armstrong, A. G., & Hagel, J. (1995). Real profits from virtual communities. McKinsey
Quarterly, 3, 128-141.
Armstrong, A., G., & Hagel, J. (1996). The real value of online communities. Harvard
Business Review, 74 (3), 134-141.
Arsal, I., Backman, S., & Baldwin, E. (2008). Influence of an online travel community on
travel decisions. Information and Communication Technologies in Tourism 2008:
Proceedings of the International Conference (pp. 82-93). Innsbruck, Austria.
_26,000_Likes_on_Facebook/
Chung, J. Y., & Buhalis, D. (2008). Information needs in online social networks.
Information Technology & Tourism, 10, 267-281.
Coon, D. A. (1998). An investigation of friends‘ internet relay chat as a community.
(Master‘s thesis), Kansas State University, Manhattan, KS.
Dellarocas, C. (2001). Building trust online: The design of reliable reputation reporting:

144
mechanisms for online trading communities. Retrieved from
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=289967
Deighton, J. (1992). The consumption of performance. Journal of Consumer Research, 19,
362-372
Gretzel, U., Kang, M., & Lee, W. J. (2008). Differences in consumer-generated
media adoption and use: A cross-national perspective. Journal of Hospitality and
Leisure Marketing, 17 (1-2), 99-120.
Gu, B., & Jarvenpaa, S. (2003). Online discussion boards for technical support: The effect of
token recognition on customer contributions. Proceedings of International
Han, H., & Kim, W. (2009). Outcomes of relational benefits: Restaurant customers'
perspective. Journal of Travel & Tourism Marketing, 26, 820–835.
Han, P., & Maclaurin, A. (2002). Do consumers really care about online privacy? Marketing
Management, 11 (1), 35-38.
Harwood, T., & Gary, T. (2009). Infiltrating an e-tribe: marketing within the Machinima
[computerized games] community. Journal of Customer Behaviour, 8 (1), 67-83.
He, Y., & Song, H. (2009). A mediation model of tourists' repurchase intentions for
packaged tour services. Retrieved from
http://repository.lib.polyu.edu.hk/jspui/bitstream/10397/1720/1/17- Mediation
%20Model.pdf.
Hernández, B., Jiménez, J., & Martin, M.J. (2011). Age, gender and income: Do they really
moderate online shopping behavior? Online Information Review, 35 (1), 113-133.
Hertzog, C., & Hultsch, D. F. (2000). Metacognition in adulthood and old age. In F.I.M.
Craik and T. A. Salthouse (Eds.). Handbook of aging and cognition (2nd ed., pp. 417-
466). Mahwah, NJ: Erlbaum.
Hess, J., & Story, J. (2005). Trust-based commitment: multidimensional consumer-brand
relationships. Journal of Consumer Marketing, 22 (6), 313-322.
Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated
environments: Conceptual foundations. Journal of Marketing, 60, 50-68.
Hogg, M. A., & Abrams, D. (1988). Social Identifications. London: Routledge