Marketing Intelligence & Planning: Luxury Restaurant Loyalty

This research paper, "Diners’ loyalty toward luxury restaurants: the moderating role of product knowledge," investigates the factors influencing consumer behavior and brand loyalty within the luxury restaurant sector. The study, based on a modified Mehrabian-Russell model, examines how restaurant stimuli such as food quality, service quality, and atmospherics impact diners' emotions and subsequent loyalty. The findings reveal that these stimuli influence diners' emotions, which in turn affect brand loyalty. Furthermore, the research highlights the direct influence of food quality on loyalty and the moderating role of diners' product knowledge in shaping the relationship between restaurant stimuli and emotion. The study utilizes data from 238 consumers who have dined at Michelin-starred luxury restaurants in Hong Kong and employs structural equation modeling to analyze the data. The paper contributes to both theoretical and managerial understanding of luxury restaurant consumption, offering implications for hospitality and brand managers. The research emphasizes the importance of product knowledge in building brand loyalty and provides insights into consumer behavior within the luxury service sector.

Marketing Intelligence & Planning
Diners’ loyalty toward luxury restaurants: the moderating role of product
knowledge
Norman Peng Annie Huiling Chen
Article information:
To cite this document:
Norman Peng Annie Huiling Chen , (2015),"Diners’ loyalty toward luxury restaurants: the moderating
role of product knowledge", Marketing Intelligence & Planning, Vol. 33 Iss 2 pp. 179 - 196
Permanent link to this document:
http://dx.doi.org/10.1108/MIP-03-2014-0049
Downloaded on: 09 February 2016, At: 06:35 (PT)
References: this document contains references to 60 other documents.
To copy this document: [e-mail]
The fulltext of this document has been downloaded 590 times since 2015*
Users who downloaded this article also downloaded:
Annie Chen, Norman Peng, Kuang-peng Hung, (2015),"The effects of luxury restaurant environments
on diners’ emotions and loyalty: Incorporating diner expectations into an extended Mehrabian-Russell
model", International Journal of Contemporary Hospitality Management, Vol. 27 Iss 2 pp. 236-260
http://dx.doi.org/10.1108/IJCHM-07-2013-0280
Kisang Ryu, Hye-Rin Lee, Woo Gon Kim, (2012),"The influence of the quality of the physical
environment, food, and service on restaurant image, customer perceived value, customer satisfaction,
and behavioral intentions", International Journal of Contemporary Hospitality Management, Vol. 24 Iss
2 pp. 200-223 http://dx.doi.org/10.1108/09596111211206141
Milos Bujisic, Joe Hutchinson, H.G. Parsa, (2014),"The effects of restaurant quality attributes on
customer behavioral intentions", International Journal of Contemporary Hospitality Management, Vol.
26 Iss 8 pp. 1270-1291 http://dx.doi.org/10.1108/IJCHM-04-2013-0162
Access to this document was granted through an Emerald subscription provided by emerald-
srm:272736 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emera
for Authors service information about how to choose which publication to write for and submis
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more informa
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The compa
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes
well as providing an extensive range of online products and additional customer resources and
services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the
Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for
digital archive preservation.
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

*Related content and download information correct at time of
download.
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

Diners’ loyalty toward luxury
restaurants: the moderating
role of product knowledge
Norman Peng
University of Westminster, London, UK, and
Annie Huiling Chen
Ming Chuan University, Taipei, Taiwan
Abstract
Purpose – Consumersdine at luxury restaurantsfor reasonsbeyond fulfilling basicneeds;
however,little is known about the factors that contribute to diners’loyalty.The purpose of this paper
is to examine diners’luxury restaurant consumption behavior by incorporating product knowledge
into a modified Mehrabian-Russell model.
Design/methodology/approach – Following exploratory qualitative research,238 consumers who
have dined at Hong Kong’s Michelin-starred luxury restaurants were recruited for the main study.
The data were analyzed through structural equation modeling.
Findings – The results show that luxury restaurants’stimuli (i.e.food quality,service quality,and
atmospherics)influence diners’emotions,which in turn affecttheirbrand loyalty.Furthermore,
food quality can directly influence diners’loyalty toward the restaurant.Third, diners’product
knowledge can moderate the relationships between restaurant stimuli and diners’emotion.
Research limitations/implications – This study offers new empirical support for the proposition
that product knowledge has a role in building brand loyalty and thereby shades both theoretical and
managerial understanding of the luxury restaurant consumption process.
Originality/value – This study is one of the first to conceptualize diners’loyalty toward luxury
restaurantsby examining the influencesof restaurants’stimuliand diners’knowledgetoward
luxury restaurants.In addition,this study puts forth some managerial implications for practitioners.
Keywords Emotion,Luxury restaurant,Product knowledge,Stimuli
Paper type Research paper
Introduction
This paper’s purpose is to examine Michelin-starred luxury restaurant diners’ consumptio
behaviorby incorporating productknowledgeinto a modified Mehrabian-Russell
model.The Mehrabian and Russell (1974) model studies environment stimuli’s effects
on individuals’emotions and subsequent behaviors.The value of the luxury goods
marketwas close to US$302[1]billion worldwide in 2012,which is more than five
times the value ofthe industry in 1997 (Nueno and Quelch,1998;Roberts,2012).
Scholars have consistently been interested in exploring the luxury goods market and
its consumers (e.g.Berthon etal.,2009;Chadha and Husband,2006;Gardyn,2002;
Jolson etal.,1981;Nueno and Quelch,1998).This interesthas been further driven
by observing the robustness ofthis issue,given the recentand ongoing global
recession (Sullivan,2009;Wood,2009).To date,scholars have suggested thatthe
appeal of luxury goods can be a result of their ability to signify luxury value to users
and others (Chadha and Husband,2006;Gardyn,2002;Jolson etal.,1981;Nueno
and Quelch,1998).This ability is mainly due to luxury goods’perceived premium
quality,recognizable style, reputation, and/or limited accessibility (Berthon et al., 2009;
Hung et al.,2011).
Marketing Intelligence & Planning
Vol.33 No.2,2015
pp.179-196
© Emerald Group Publishing Limited
0263-4503
DOI 10.1108/MIP-03-2014-0049
Received 16 March 201
Revised 16 March 2014
Accepted 9 January 201
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-4503.htm
179
Diners’loyalty
toward luxury
restaurants
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

Although progress has been made,Berthon etal.(2009)note that luxury goods
consumption behavior is still poorly understood and under investigated.Furthermore,
the focus of existing studies has been mainly on the consumption of tangible lu
goods,such as handbags,accessories,and cars (e.g.Han et al.,2010a, b;Hung et al.,
2011);few studies have examined the consumption of luxury services,even though
luxury products have become more accessible to middle-class consumers since
mid-2000s (Lee and Hwang, 2011; Mintel, 2011; Silverstein and Fiske, 2003). B
the gaps in the literature,relatively little is known about luxury services consumptio
behavior and the factors influencing this behavior.
Within the range of luxury services,restaurant industry has its significances. First,
the restaurantindustry has grown significantly since the 1980s.From the 1980s
to 2011,restaurant industry sales increased more than ten times,from $42.8 billion to
$536.7 billion,worldwide.Second,according to Wu and Liang (2009)and Lee and
Hwang (2011),fine dining is a luxury leisure activity with additionalimportance.
Apart from fine dining being increasingly valued by middle-class consumers an
families,it is also worth investigating further whether existing frameworks on lux
goods consumption are applicable to service-based products,given such products’
perishable and intangible nature.
To narrow the gaps in the luxury consumption literature, this study examines
Kong diners’loyalty toward luxury restaurantsby considering Michelin-starred
restaurants’stimuli(i.e.food quality,service quality,and atmospherics)and diners’
productknowledge.The objectives of this study are as follows.The first objective
is to conceptualize customers’ loyalty toward Michelin-starred luxury restauran
a modified Mehrabian-Russellmodel.The second objective is to test the moderating
effectof diners’productknowledge.Finally,some implications for hospitality and
brand managers are discussed.
Literature review
Luxury restaurants – definition and previous studies
Among different types of restaurants (e.g.full/limited-service and fine/family/casual
dining environments),full-serviceand fine dining restaurantshave received
considerable attention since 2000 (Lee and Hwang,2011).Wu and Liang’s (2009)
research clearly defines luxury restaurants.In their study,four-and five-star hotels’
full-service restaurants are considered luxury restaurants because they are sub
stringent inspection,high customer expectations,and competition from nearby luxury
hotels and restaurants. In addition to Wu and Liang’s definition, Lee and Hwang
suggestthat luxury restaurantshave the following characteristics:full-service,
a high-quality environment,and entrées that are generally priced over $20.Based on
the developmentof the restaurantindustry’s and the definitions ofluxury goods
proposed by Berthon etal. (2009),Han et al. (2010a, b),Lee and Hwang (2011),
and Nueno and Quelch (1998),a luxury restaurant is considered in this study to be a
full-service restaurant whose environment (e.g.décor,atmospherics,and services) and
products (e.g.food and beverages)are carefully prepared and presented,unique,
superior in quality,and conspicuous.
In the previous literature on the behavior of consumers as restaurant patronthe
studies by Jang and Namkung (2009),Kwun and Oh (2006),and Wu and Liang (2009)
are significant.First,in Kwun and Oh’s (2006)study,the authors investigate the
impact of restaurants’performance factors on new and experienced customers.Their
results indicate how perceived performance (i.e.service,food,beverage,facilities,and
180
MIP
33,2
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

atmospherics)influencesnew and existingcustomers’revisit intentionsand
recommendation behavior differently.Kwun and Oh’s research has contributed to
the literature, but their paper’s relationship to existing leisure and/or hospitality-related
theory has not been discussed.In addition,they focus on full-service restaurants,not
luxury restaurants.Second,Wu and Liang (2009) contribute to the luxury restaurant
consumption research literature by focussing on service performance,environmental
stimulus,and additionalinfluenceson diners’perceived experientialvalueand
satisfaction. Their results show that restaurant staff, the dining environment, and other
dinerscan impactcustomers’perceived experientialvalueof luxury restaurants
and their satisfaction with their dining experience.Although Wu and Liang’s (2009)
research is significant,it is not without limitations.Wu and Liang (2009)examined
diners’levelof satisfaction;however,they did notconsider the impactof diners’
knowledge aboutrestaurants and diners’future behavioralintentions,such as the
intention to revisit.
Finally,Jang and Namkung’s (2009) study had the most direct implications for the
presentstudy’s framework.Using a modified Mehrabian-Russellmodelto examine
restaurant-specific stimuli’s (i.e.productquality,atmospherics,and service quality)
impact on middle-class diners’emotion,their results showed that service quality and
atmosphericshavepositiveimpactson diners’positiveemotionswhile product
quality has a negative correlation with diners’negative emotions.Last but not least,
their results show thatonly positive emotions affectdiners’subsequentbehavioral
intentions.Althoughthe Mehrabian-Russellmodelis often used to examine
service-related consumption,Jang and Namkung’s findings and framework cannot
be applied to the luxury restaurant consumption scenario without considering luxury
and other restaurants’differences.Mainly,apartfrom fulfilling basic needs,diners
often go to luxury restaurants for purposes that are different than when they visit
family and/orcasualrestaurants(Leeand Hwang,2011;Wu and Liang,2009).
Because Jang and Namkung’s research context is general full-service restaurants, their
modeldoes nottake diners’understanding aboutluxury restaurants into account.
Based on this review ofthe currentliterature on luxury restaurantconsumption,
several research opportunities have emerged.
Research framework and hypotheses
Mehrabian and Russell’s (1974)study on the influences of environment stimuliand
the modelthey proposed had a significantimpacton the study ofenvironmental
psychology.According tothem,environmentstimuli(S) can affectconsumers’
emotions (O), which in turn influence their responses (R). The Mehrabian-Russell model
and its modifications have been applied in retail, service, leisure, and tourism contexts.
The contexts thathave been examined include shopping (Ong and Khong,2011;
Tai and Fung,1997),retailmarketing (Kaltcheva and Weitz,2006),online shopping
(Koo and Ju, 2010),tourists’shopping behavior(Yuksel,2007),and restaurant
management(Jang and Namkung,2009).These studies’findings generally support
the notion thatstimuliaffectconsumers’emotions,which in turn influence their
responses (e.g.revisiting and/or recommending).
After examining these studies and considering the contexts of their research,the
models ofJang and Namkung (2009)were adopted,with modifications,for use in
this study.This currentresearch adopted a modified version oftheirframework
because ofits comprehensive nature:theirframework considers the influence of
multiple stimuli.In addition to environmental stimuli,service quality and food quality
181
Diners’loyalty
toward luxury
restaurants
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

are included in their model when examining diners’ emotions and subsequent b
This study’sframework doesnot include“negativeemotion”becauseJang and
Namkung (2009) found that this variable did not influence diners’ behavior and
influenced by a restaurant’s service quality or atmosphere.Therefore,it seems that
“negative emotion” has limited impact when the Mehrabian-Russellmodelis used to
examinediners’behavior.In addition,severalscholarswho have used the
Mehrabian-Russellmodelto study service consumption scenarios did notinclude
negative emotion in their framework (e.g.Koo and Ju,2010;Lee and Hwang,2011;
Tai and Fung,1997;Yuksel,2007).
In addition to the variables mentioned above,product knowledge is included in the
present framework as a new moderating variable that can affect luxury restaur
diners’consumption behavior (Berthon et al.,2009;Chen and Peng,2012;Wong and
Yeh,2009).The following section provides the details and definitions of each varia
followed by this study’s proposed hypotheses (Figure 1).
According to Jang and Namkung (2009),Kwun and Oh (2006),and Wu and Liang
(2009),a restaurant’s stimuli influence diners’emotions.Consumption emotions can be
described as the affective responses generated specifically during product usag
consumption experiences (Han et al., 2009; King and Meiselman, 2010; King et
When examining customers’emotions,questions thatseek to determine whether
respondents are excited and/or pleased are often included.
Koo and Ju (2010)suggest that during service encounters,environmentalstimuli
affectcustomers’emotions,which in turn influencetheir behavioralintentions.
According to Jang and Namkung (2009),Kwun and Oh (2006),and Wu and Liang
(2009),a restaurant’senvironmentincludesservicequality,food quality,and
atmospherics.Kwun and Oh (2006)define food and beverages as restaurants’core
products.The freshnessand tastinessof food can havesignificantimpactson
consumers’emotions.In addition,Wu and Liang (2009)note thatthe presentation
H5a
H5cH3
H1
H2
Atmospherics
Food quality
Service
quality
H5d
H4 Brand loyaltyEmotion
Product
knowledge
H5b
Figure 1.
Research framework
182
MIP
33,2
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

of food and beverages is an important part of luxury restaurants’ food quality and can
influence diners’ emotions.Based on the findings of previous literature (e.g.Kwun and
Oh,2006;Wu and Liang,2009),this research hypothesizes the following:
H1. Diners’perceptions of luxury restaurants’food quality has a positive effect on
their emotions.
In the contextof fine dining,restaurants’service personnelinteractwith diners
throughout their visit (Jang and Namkung,2009;Kwun and Oh,2006;Wu and Liang,
2009).Service quality is customers’overallperception ofthe relative inferiority or
superiority ofa service provider(Parasuraman etal.,1988).In the contextof a
restaurant,the staff’s knowledge aboutthe restaurantand its menu,their levelof
helpfulness,and whetherthey are reliable ornot are often considered.Previous
literature has confirmed that staffs’service quality affects diners’emotions (e.g.Jang
and Namkung,2009;Kwun and Oh,2006);hence,this study proposes that customers’
perceptions of service quality have positive effects on their emotions:
H2. Diners’perceptions of luxury restaurants’service quality have positive effects
on their emotions.
Third, based on studies ofretailshops,convention centers,and shopping malls,
Breiter and Milman (2006),Kaltcheva and Weitz (2006),Koo and Ju (2010),Ong et al.
(2012),and Taiand Fung (1997)confirm that atmospherics has a profound impact
on customers’experiencesand subsequentbehavior.Kotler (1973)defines
atmospherics as the conscious design of space to creating certain buyer effects.The
main aim is to increasepurchaseprobability.Key techniquesincludeapplying
suitable color,lighting,music,and décor to a service environment (Koo and Ju,2010;
Ong et al.,2012;Tai and Fung,1997).In the context of a restaurant’s,Liu and Jang
(2009)find that atmosphericscan affect diners’emotions;hence,this study
hypothesizes the following:
H3. Diners’ perceptions of luxury restaurants’ atmospherics have positive effects on
their emotions.
Previous literature on consumer behavior suggests that customers’positive emotions
will make them loyal,e.g.,likely to revisit in the future (Flavian et al.,2001;Jeon and
Hyun,2012;Yoon and Uysal,2005).According to Oliver (1999),brand loyalty can be
defined as a deeply held commitment to repurchase or repatronize a preferred product
or service consistently in the future.In this study’s context,brand loyalty refers to
diners’intentions to repeatedly revisitthe same restaurant.In studying restaurant
customers and casino patrons,Jang and Namkung (2009)and Jeon and Hyun (2012)
find that positive emotion or being satisfied with the product can lead to significantly
strongerbrand loyalty.In otherwords,customers with strong positive emotions
are more likely to become repeat customers.Based on the above-mentioned literature,
this study examines the following hypothesis:
H4. Diners’emotions have positive effects on their loyalty toward restaurants.
In addition to the variables mentioned above,this research incorporates product
knowledge into its proposed framework.According to Laroche et al. (2001),knowledge
influences allphases of the consumption process.In particular,productknowledge
can significantly influence consumers when they process (e.g. noticing and evaluating)
market/productinformation (Lin and Chen,2006).Productknowledgehas been
183
Diners’loyalty
toward luxury
restaurants
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

investigated when scholars research the consumption ofluxury goods (Bian and
Moutinho,2011) and leisure activities (Chen and Peng,2012).
Scholars have generally agreed productknowledge is an importantfactor when
examining consumption behaviors;nonetheless,its moderating effect received lesser
attention.In previous studies of hospitality services,Chen and Peng (2012) confirmed
that lodgers’knowledge regarding hospitality products moderates their consumpt
processes.They find thatcustomers are more willing to stay atenvironmentally
friendly hotels if they consider themselves to be knowledgeable about the deta
green issues in generaland green products in particular.Additionally,in Wong
and Yeh’s (2009)research on tourists’behavior,they confirmed thattourists’
knowledge can significantly moderate tourists’decision-making processes.
Productknowledge’s moderating effecthas been examined before;nevertheless,
this research aims to further explore how high productknowledge diners and low
productknowledge diners behave when exposed to luxury restaurants’stimuli.To
date,no known study has examined how restaurantstimuli’s influences on diners
are moderated by product knowledge.For Fryxell and Lo (2003),consumers’product
knowledge covers general knowledge of facts,concepts,and relationships concerning
the product in question. A question such as “Compared to my friends/average p
know more aboutthis product”is often used to assess customers’self-perceived
knowledgeabouta given product.Within thecontextof this research context,
restaurants’stimuli(i.e.food quality,service quality,and atmospherics)make up a
restaurant’soverallproduct(Jang and Namkung,2009).Based on theliterature
discussed above,this research proposes H5:
H5. Diners’productknowledge(i.e.high or low) can moderatetheir luxury
restaurant consumption process. Specifically, first, food quality have a str
positive relationship with diners’emotions;second,service quality have a
stronger positive relationship with diners’ emotions; third, atmospherics h
strongerpositiverelationship with diners’emotions;and fourth,diners’
emotionshavea strongerpositiverelationship with theirloyalty toward
restaurants for diners with higher a product knowledge vs diners with a lo
product knowledge.
Methodology
Qualitative pretests
Prior to the main study,qualitative research in the form of interviews was conducte
for exploratory purposes. The purpose of this exploratory research was to impr
reliability and validity ofthe main quantitative research.In addition,it aimed to
increase the generalizability of the main research and reflect diners’actualemotions
and subsequent behavioral intentions.Eight in-depth interviews were conducted.The
area ofthe sampling was within Hong Kong’s main island.A purposive sampling
method was used.All participants had dined at a five-star,full-service luxury hotel
restaurantin Hong Kong atleastonce.According to Wu and Liang (2009),these
restaurants are suitable as they are subjectto stringentinspection,high customer
expectations, and competition from nearby luxury hotels and restaurants. At th
a luxury restaurantwas defined as a full-service restaurantwith a high-quality
environment and a cost of $20 or more for most entrées (Lee and Hwang,2011).
Each interview lasted approximately 30-45 minutes.The interviews were audio
recorded in Mandarin and/or Cantonese and transcribed afterward.The objectives of
184
MIP
33,2
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

studying these interviews were to:first,explore the factors that contributed to diners’
decisions to dine at luxury restaurants;second,formulate an operational definition in
the contextof this research (e.g.the definition ofluxury restaurantand product
knowledge in the context of luxury restaurant); and third, provide comments on how to
improve the quality of the main study’s survey questions.
The interview sessions consisted of two stages. During the first part of each session,
interviewees shared their luxury restaurant dining experiences with the interviewer
and focussed on the factors that influenced their emotions.A set of semi-structured
questions was prepared based on the findings ofprevious studies.The transcripts
were analyzed using an interpretive approach.
By examining the participants’responses,the study’s framework could be used
to describe theiremotionsand behavioralintentionsappropriately.Nevertheless,
this study’s definition ofa luxury restaurantneeded modifications.Based on the
respondents’feedback,the operational definition of a luxury restaurant in Hong Kong
was revised to have the following characteristics: first, a full-service restaurant that has
one or more Michelin star(s);second,main courses with an average price above $30;
third,a set meal’s average price above $67 (excluding service charge and tips).
Additionally,participants suggested that only diners who dined for leisure purposes
should be included in the main quantitative study. The above procedures were adopted
afterconsidering the research ofHung etal. (2011),Jang and Namkung (2009),
Lee and Hwang (2011),and Wu and Liang (2009).
During the second half of the interviews,copies of the main study’s survey were
shown to the pilot study respondents.This was done to improve the main study’s
quality.Participants commented on issues such as the clarity of the questions,the
layout of the survey,and redundant questions.Because this stage’s objective was to
refine the main study’s questionnaire,it was more structured in terms of proceedings
and content than the first stage.
Sampling and data collection methods of the quantitative studies
Once the survey was finalized,238 participants were recruited through an on-site
sampling method. The data used for this study were collected at 14 luxury restaurants
in Hong Kong.Based on information provided by the Michelin Guide Hong Kong and
Macau and theweb sitewww.tripadvisor.com (Tripadvisor,2013),thereare 61
Michelin-starred restaurantsin Hong Kong. The interview took placenearthe
restaurants or hotels (for restaurants in hotels) to increase the probability of meeting
participants who had just finished dining at one of the luxury restaurants.Based on
King et al.’s (2010) recommendation,the most appropriate time to evaluate emotions is
during the exposure to the stimulus or immediately after the exposure. The restaurants
included in this research were mainly serving Chinese (including local dishes),French,
Italian,Japanese,and fusion dishes.
To qualify for the interview,potential participants had to:be above the age of 18;
have dined at Michelin-starred luxury restaurants for leisure purposes before;be a
residents of Hong Kong. Before the interview, the interviewers explained the purpose of
the research and the definition ofa luxury restaurantused in this study.Of the
interviews conducted,238 results were deemed to be effective,resulting in a valid
return rate of71 percent.The participants were primarily female (50.8 percent),
working in the service industry (28.7 percent),and between the ages of36 and 40
(29.4 percent).
185
Diners’loyalty
toward luxury
restaurants
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

Questionnaires used in the main study
The participants completed a 19-question survey that evaluated diners’ loyalty
luxury restaurants.This study adopted existing validated scalesto assessits
constructs (Chen and Peng, 2012; Hung et al., 2011;Jang and Namkung, 2009;Jeon and
Hyun,2012;Lee and Hwang,2011;Wu and Liang,2009).The target research question
under examination was the following:
RQ1. How will luxury restaurants’ stimuli and diners’ product knowledge influ
their restaurant consumption behavior?
All of the variables in the model(Figure 1)were measured with multiple items.In
addition,they were all found to be reliable,with values of Cronbach’s α ranging from
0.81 to 0.96. A seven-point Likert-type scale was used when designing the itemSome
of the items were rephrased to maintain consistency.The items for each variable are
presented in Table I.
Data analysis
Analyticalapproach, data screening, and the measurement model
SPSS and AMOS 7.0 were used to analyze the data. Following the two-step app
Anderson and Gerbing (1988),a measurementmodelwas first estimated using
confirmatory factoranalysis(CFA). After assessmentof the adequacy ofthe
Variable Measurement items SFL α CR AVE
Food quality (FQ)FQ1:food presentation is visually attractive 0.730.83 0.840.63
FQ2:the restaurant serves tasty food 0.81
FQ3:the restaurant offers fresh food 0.84
Service quality
(SQ)
SQ1:the staffs provide a thorough and satisfactory
service
0.750.82 0.870.69
SQ2:the staffs are reliable 0.90
SQ3:the staffs are professional 0.83
Atmospherics (A)A1:restaurant’s decoration is impressive 0.750.86 0.820.61
A2:restaurant’s interior design is appealing 0.86
A3:restaurant’s atmosphere is luxurious 0.72
Emotion (E) I feel … not at all (1) – strongly (7) 0.93 0.930.81
E1:joy (joyful,pleased,romantic,welcoming) 0.92
E2:excitement (excited,thrilled,enthusiastic) 0.95
E3:peacefulness (comfortable,relaxed,at rest) 0.83
Product
knowledge (PK)
PK1:compared to average person,I am familiar
with luxury restaurants’environment
0.870.92 0.920.80
PK2:compared to my friends,I am familiar with
luxury restaurants’offers
0.86
PK3:compared to people who dine at restaurants
a lot,I am familiar with luxury restaurants
0.95
Brand loyalty (BL)BL1:I would like to come back to this restaurant
in the future
0.940.97 0.970.87
BL2:I plan to revisit this restaurant in the future 0.95
BL3:I intend to continue to be this restaurant’s customer0.93
BL4:I will visit this restaurant when I dine at luxury
restaurant next time
0.93
Notes: SFL,Standardized factor loadings;α,Cronbach’s α;CR,composite reliability;AVE, average
variance extracted.aSome of the items were rephrased in order to maintain consistency
Table I.
Descriptive analysis
of the measures
186
MIP
33,2
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

measurementmodel,structuralequation modeling wasutilized toidentify the
best-fitting modeland to examine the causalrelationships.CFA, using maximum
likelihood estimation with the 238 casesgathered,was conducted to assessthe
structure of the variables in the model proposed in this research.After the model was
assessedfor unidimensionality,reliability,and constructvalidity,the results
demonstrated that the modelfit the data well.Nevertheless,the standardized factor
loading value for one of the emotion items did not meet the minimum value of 0.40,
and thus, this item was eliminated to enhance reliability and decrease error (Ford et al.,
1986).After this item wasexcluded,the remaining itemswereagain subjected
to CFA,and the results showed a satisfactory fitto the data (χ2¼ 202.39,df ¼ 94,
p o 0.001,RMSEA ¼ 0.07,CFI ¼ 0.966,NFI ¼ 0.938).
The measurementmodel consistedof five multiple-item latentvariables
(i.e.food quality,service quality,atmospherics,emotion,and brand loyalty).Each
constructwas then tested by CFA and showed a good modelfit. The composite
reliability (CR)of the constructwas used to measure the latentvariable’s internal
consistency.The CR values of this research’s constructs were between 0.82 and 0.97.
These values are greater than the value of 0.60 recommend by Fornelland Larcker
(1981) and Bagozzi and Yi (1988).Furthermore,the average variance extracted (AVE)
was calculated. Because all the AVE values, ranging from 0.61 to 0.87, exceeded Fornell
and Larcker’s (1981)recommended value of 0.50,convergent validity was confirmed.
Last but not least, discriminant validity was judged to have been achieved because the
AVE value foreach constructwas greaterthan the squared correlation between
constructs (Table II). This result indicates that constructs with different purposes have
little relation to one another.After CFA was performed,based on Bagozzi’s (1983),
Kline’s(2005),and Thompson (2004)recommendations,this research’svariables,
reliability, convergent validity, and discriminant validity were judged to be acceptable;
therefore,the items were considered suitable for use in determining the average of
each factor.
Structuralmodel
The results gathered afterusing structuralequation modeling showed a good fit
(χ2¼ 242.296,df ¼ 97,χ2/df ¼ 2.498;p o 0.001,RMSEA ¼ 0.079,CFI ¼ 0.954).In other
words, the results of this study provide support for its proposed framework. As for the
hypotheses,H1 was supported,with a structuralestimate of 0.43 (t ¼ 2.35,p o 0.01).
This means thata restaurant’s food quality has a significantimpacton diners’
emotions.H2 was supported,with a structural estimate of 0.41 (t ¼ 2.93,p o 0.01).In
other words,service quality positively influences diners’emotions.H3 was supported,
with a structuralestimate of 0.24 (t ¼ 2.16,p o 0.01).This means that atmospherics
has a significant and positive impact on diners’emotions.Finally,H4 was supported,
FQ SQ A E BL
Food quality (FQ) 0.63
Service quality (SQ) 0.66 0.69
Atmospherics (A) 0.77 0.61 0.61
Emotion (E) 0.62 0.54 0.63 0.81
Brand loyalty (BL) 0.54 0.37 0.45 0.33 0.87
Notes: Italic numbers on the diagonalparentheses are square root of each construct’s AVE value.
**p o0.01
Table II.
Correlation between
constructs following
CFA
187
Diners’loyalty
toward luxury
restaurants
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)

with a structural estimate of 0.37 (t ¼ 5.38,p o 0.001).This means that emotions affect
diners’brand loyalty positively (Table III).
Mediating effect
A Sobel test was performed to determinethe mediating effectsof emotion on
restaurants’stimuli(i.e.food quality,service quality,and atmospherics)and brand
loyalty (Sobel,1982).The results (i.e. Z W 1.96) indicate that emotion fully mediates
relationships between service quality and brand loyalty (Z ¼ 2.57) and the rela
between atmospherics and brand loyalty (Z ¼ 2.01).Hence,emotion fully mediates
these two relationships.However,emotion is a partialmediator in the relationship
between food quality and brand loyalty (Z ¼ 2.19).
Moderating effect
To testthe hypothesized moderating effects ofproductknowledge,an invariance
analysisof differentgroupswas applied (Jurowskiand Gursoy,2004),and the
procedure recommended by Han et al. (2010a, b) and Bell and Menguc (2002)
The participants were divided into two groups (i.e.high and low product knowledge)
based on their survey responses.Initially,the structural models for the high and low
productknowledgegroupswereestimated withoutacross-group constraints(i.e.
unconstrained models).Then,across-group constraints (i.e.constrained model) where
the parameter estimates for high and low product knowledge groups were cons
to be equal.Finally,a χ2 test comparing the unconstrained and constrained models
was used to detectmoderating effects.A significantχ2 differencebetween the
unconstrained models suggests that there are some differences between diner
high and low productknowledgein terms of luxury restaurantconsumption
behavior. The findings show H5a and H5c were supported, but H5b and H5d we
supported (Table IV).
Path Standard estimate t Test
H1:FQ→E 0.43 2.35** Support
H2:SQ→E 0.41 2.93** Support
H3:A→E 0.24 2.16** Support
H4:E→BL 0.37 5.38*** Support
Notes: FQ, Food quality;SQ, service quality;A, atmospherics;E, emotion;BL, brand loyalty.
*p o 0.05;**p o 0.01;***p o 0.001
Table III.
Hypotheses tests
High product
knowledge group
(n ¼ 128)
Low product
knowledge group
(n ¼ 110)
Path
Results
(significance level)
Results
(significance level)
Product
knowledge’s
moderating effect
(H5)
Food quality→emotion Support** Reject Significant (H5a)
Service quality→emotion Support** Support** Reject (H5b)
Atmospherics→emotion Support** Reject Significant (H5c)
Emotion→brand loyalty Support*** Support*** Reject (H5d)
Notes: *p o 0.05;**p o 0.01;***p o 0.001
Table IV.
Testing the
moderating effect of
product knowledge
188
MIP
33,2
Downloaded by FLINDERS UNIVERSITY OF SOUTH AUSTRALIA At 06:35 09 February 2016 (PT)