See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/259683378

Service Quality Perception and Satisfaction:

Buying Behaviour Prediction in an Australian
Festivalscape

Article in International Journal of Tourism Research January 2014

DOI: 10.1002/jtr.1901

CITATIONS READS

19 282

1 author:

Johan Bruwer
University of South Australia
102 PUBLICATIONS 1,664 CITATIONS

SEE PROFILE

All content following this page was uploaded by Johan Bruwer on 16 October 2014.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
International Journal of Tourism Research, Int. J. Tourism Res., 16: 7686 (2014)
Published online 27 July 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/jtr.1901

Service Quality Perception and Satisfaction: Buying Behaviour Prediction in

an Australian Festivalscape
JOHAN BRUWER*
School of Agriculture, Food and Wine, The University of Adelaide, Adelaide, South Australia Australia

ABSTRACT

The study was conducted on 358 attendees at a major wine festival in Australia. A positive relationship between quality perception and
overall satisfaction constructs exists. New insight to festivalscape knowledge is provided through the rsttime and repeat visitor dynamic
as predictor of actual buying behaviour. Higher percentage of repeat visitors correlates with higher likelihood of (wine) buying. Overall sat-
isfaction is a stronger predictor of buying behaviour than any individual service quality dimension and of these quality dimensions overall.
Repeat visitors, 35 years and older, are the highest yielding visitor group from a nancial viewpoint. Firsttime visitors are more shortterm
oriented in their planning when making the nal decision to attend the event. Copyright 2012 John Wiley & Sons, Ltd.

Received 12 February 2012; Revised 23 May 2012; Accepted 11 June 2012

key words festivalscape; quality perception; satisfaction; behavioural intention; servicescape; repeat visitor; rsttime visitor; wine

INTRODUCTION improved considerably over the past decade or two (Getz,

Traditionally, wine tourism research was focused on what
motivated tourists to taste and buy wine when visiting
regional wine destinations (Cohen and BenNun, 2009).
Most often, these destinations were wineries and, more spe-
cically, their tasting rooms where direct interaction with
the tourists took place. Special events in the form of festivals,
whether themed or otherwise, received little attention from
researchers within the ambit of wine tourism research.
The profound benets a festival have for their local
communities have been widely touted, mainly in an eco-
nomic sense. The supposed benecial economic impact on
the host community surrounding the festival area due to
visitor expenditures is one of the most important reasons for
arranging a festival in the rst place (Thrane, 2002). Not sur-
prising, there has been a sharply increased focus in the litera-
ture on festivals, and most of the focus has been on the
economic benets (Jackson et al., 2005; Yoon et al., 2010).
Festivals are also used as a means of destination branding,
and there tends to be a big difference between smaller regional
festivals (e.g. a wine festival) and national festival mega
events held in the metropolitan areas. The difference is not
only in size but also in the underlying structure in terms of
local community involvement (Lee and Arcodia, 2011). In
the words of Derrett (2003, p.38), festivals and events
demonstrate the popular denitions of a sense of community
through offering connections, belonging, support, empower-
ment, participation and safety. Festivals are events generally
short in duration, with a specic theme, creating a bundle of
experiences that appeal to attendees (Saleh and Ryan, 1993).
A recent analysis of the nature and scope of 423 festival
research studies concluded that the scholarship level has
*Correspondence to: Professor Johan Bruwer, School of Agriculture, Food
and Wine, The University of Adelaide, Waite Campus, PMB 1, Glen
Osmond, South Australia 5064, Australia.
Email: johan.bruwer@adelaide.edu.au
2010). Three of the identied research themes are relevant
to the current study, namely, evaluations (36 citations),
economic impact (132 citations) and marketing (57 citations).
Often these, themes are also interrelated. The majority of the
studies categorized as evaluations assessed quality or satis-
faction (Getz, 2010), which is what this study also does.
LITERATURE REVIEW
Festivals held within wine regions form part of the wine
tourism product offering and attract visitors to the area.
Festivalscape within the servicescape
Servicescape theory (Bitner, 1992) contends that the physical
environment in which a service response is experienced
affects the perception of service quality and satisfaction.
The servicescape context has been likened to what has been
described as the festivalscape (Lee et al., 2008). A
festivalscapeoriented denition of the servicescape describes
it as the general atmosphere experienced by festival patrons
(Lee et al., 2008: p.57).
Service quality of festivals
There is little agreement as to how to best dene quality, but
Cunnell and Prentice (2000) point to its importance because
of rising competition, increased consumer choice and other
forces. They distinguish between researcherdened and con-
sumerdened conceptualizations thereof. The literature base
(Getz, 2010) reveals that researcherdened conceptualiza-
tions are by far the most common of the two approaches.
Our study also adopted that approach.
It has been shown that both service quality and visitor
satisfaction inuence the future behavioural intentions of fes-
tival visitors (Baker and Crompton, 2000; Cole et al., 2002;
Lee et al., 2007), whereas Cole and Illum (2006) found that

Copyright 2012 John Wiley & Sons, Ltd.


experience quality has a direct impact on visitors' future
behavioural intentions.
Performance quality is the attributes of a service con-
trolled by a tourism supplier (Baker and Crompton, 2000)
within the festivalscape of its servicescape environment.
Satisfaction, on the other hand, refers to the tourist's emo-
tional state after exposure to the opportunity (Baker and
Crompton, 2000, p.785). The quality of the opportunity to
experience the attributes of the service can be controlled
and manipulated by the service provider (Lee et al., 2007),
e.g. the wine festival provider and management committee.
The debate on the conceptualization of the performance
quality and satisfaction constructs rages on, as it does on
the nature of their interrelationships (Baker and Crompton,
2000). There is nevertheless widespread acceptance of the
need to measure service quality and at least some consensus
on two aspects of its conceptualization. It being that service
quality relates to tourists' perceptions of the (festival)
provider's performance and that their perceptions of the qual-
ity of the experience dene the satisfaction construct
(Childress and Crompton, 1997). Not surprising, studies
(36) to evaluate service quality and performance abound in
the festival research eld as determined in a recent macro
analysis by Getz (2010).
Baker and Crompton (2000) used four dimensions,
namely, generic features of the festival, specic entertain-
ment features, information sources and comfort amenities,
to measure the relationship between quality and satisfaction.
This conceptual platform has been used by several other
researchers in studies to evaluate the quality perception of
festival attendees (Lee et al., 2007; Lee et al., 2008; Cole
and Chancellor, 2009).
Service satisfaction and festivals
It is important to make a distinction between service quality
and service satisfaction. Quality perception is the cognitive
response to a service experience, whereas satisfaction is the
affective response to it (Petrick, 2004). Satisfaction has been
dened as the summary psychological state resulting when
the emotion surrounding disconrmed expectations is
coupled with the consumer's prior feelings about the
consumption experience (Oliver, 1981, p.27). In our study,
we operationalized festival satisfaction as overall satisfac-
tion, analogous to the denition of Anderson et al. (1994,
p.54) who dened it as an overall evaluation based on the
total purchase and consumption experience with a good or
service over time. An attendee's overall satisfaction with a
festival is therefore more of a holistic attitude towards the
event (Cole et al., 2002).
Although there is broad consensus that service quality and
satisfaction are different constructs, there appears to be little
agreement on the nature of their relationship (Cole and
Crompton, 2003). It should be noted that because dening
the relationship between service quality and satisfaction
was not per se the core focus of our study, we
operationalized satisfaction as overall satisfaction and only
explored its relationship with the service quality dimensions
and buying behaviour.
Copyright 2012 John Wiley & Sons, Ltd.
Festivalscape: Buying Behaviour Prediction 77
Behavioural intentions versus actual behaviour
The services marketing literature identied the roles of
service quality perception and satisfaction in the formation
of behavioural intentions. The mediating role of satisfaction
in the relationship between service quality and behavioural
intentions has been conrmed (Cole and Crompton, 2003).
Zeithaml et al. (1996) contended that behavioural intentions
are desirable behaviour or actions that visitors anticipate
they will exhibit in the future and developed a 13item scale
to measure behavioural intention. Baker and Crompton
(2000) narrowed the scale by Zeithaml et al. (1996) down
to seven items.
Previous wine festival research measured behavioural
intentions in terms of intention to revisit the festival (Cole
and Illum, 2006), likelihood of visiting local wineries and
buying locally produced wines (Yuan and Jang, 2008) and
positive wordofmouth (WOM) (Cole and Illum, 2006).
None of these studies attempted to measure whether
perceived service quality and satisfaction play a role in the
actual behaviour of festival attendees, specically their buy-
ing of wine at the festival. Our study attempted to determine
whether service quality perception and overall satisfaction
inuence wine buying and whether previous festival atten-
dance has a moderating effect on it.
Wine festivals
Wine tourism as a special interest element of tourism is not
yet greatly endowed with indepth theory (Alant and Bruwer,
2010). Wine festivals have been to an even lesser extent the
focus of targeted research studies. Wine festivals are
transient and short in duration (Yuan et al., 2008), which
provides further rationale for focusing research on the assess-
ment of service (performance) quality of various elements of
the festivalscape.
One can say that a wine festival is a wine featuring
festival that really highlights regional wines and/or has
winethemed activities and programmes (adapted from
Lee and Arcodia, 2011, p.357). These festivals have
become a popular way for towns, both large and small, to
express their unique character and distinctiveness (Lee
and Arcodia, 2011).
Wine festivals or events are important marketing tools for
both the region where it is situated and the individual winer-
ies to create brand awareness and loyalty, educate, entertain
and encourage future visitation and wine sales. (Houghton,
2001; Bruwer, 2002, 2003; Getz and Cheyne, 2002; Yuan
et al., 2006).
In some cases, the motivations for engaging in wine
tourism will revolve around the destination hub, wherein
the wine region is the primary motivational factor (Alant
and Bruwer, 2004). In other cases, the motivations will
revolve around the activity hub (Macionis, 1996), e.g. a
wine and food festival event held in a wine region but
where the event itself is the primary attraction on that
specic occasion (Bruwer, 2002). Telfer (2000) highlighted
that wine festivals provide additional sources of income for
wine regions in North America, a fact which is reiterated by
our study.
Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
78 J. Bruwer
Firsttime and repeat festival visitors
The rsttime and repeat visitor dynamic plays an impor-
tant role in the consumption of the wine tourism product
(Bruwer et al., 2012). Tourists to a destination consist of
both rsttime and repeat visitors and their visit decision
is inuenced by a number of antecedents (Um et al.,
2006). The incidence of a high percentage of rsttime
or repeat visitation in wine tourism could, in some
instances, be attributed to the spatial relationship (or lack
thereof) of the region with a big source market, as well as
through productrelated experiences (Dodd, 1999). A high
incidence of repeat visitation in wine tourism has been
conrmed in diverse recent studies in the USA
(Carmichael, 2005), Israel (Jaffe and Pasternak, 2004)
and Australia (Bruwer, 2002). The need to distinguish
between rsttime and repeat visitors to wine festivals
has also been emphasized (Shanka and Taylor, 2004).
There is general agreement that a high level of service
quality will result in satised festival attendees who are
then more likely to be communicators of positive WOM
and become repeat visitors (Cole and Illum, 2006; Cole
and Chancellor, 2009).
Studies by Shanka and Taylor (2004) and Houghton
(2001) note that, although many annual festivals rely on
repeat visitors, it is worthwhile identifying and analysing
the attributes signicantly distinguishing between rsttime
and repeat visitors. The rsttime and repeat festival visitor
dynamic is therefore relevant and will be further explored in
this study.
Following the discussion of the relevant constructs, Figure 1
represents a conceptual model of the interrelationships between
all the discussed aspects in the servicescape of a wine
festivalscape.
To operationalize the assessment of service quality in the
festivalscape environment presents an ongoing challenge for
researchers. In our study, we used a threestep approach,
which included assessing the literature base, interviewing
key informants and consulting with the festival organizers.
Following this, a symmetrical Likert scale consisting of 17
service quality and 3 satisfaction items was designed. The
study's aim was to examine the link between perceived festi-
val service quality, satisfaction, buying wine at the event and
the rsttime/repeat visitor dynamic.
First-time
Visitor
Repeat
Visitor
Figure 1. Conceptual model of interrelationships between visitor type, perceived quality, satisfaction and buying behaviour at a wine festival.
Copyright 2012 John Wiley & Sons, Ltd.
METHODOLOGY
The study was conducted at the Winery Walkabout
Festival held annually in June during the long weekend
to celebrate the birthday of Britain's Queen Elizabeth II
in the Rutherglen Wine Region of North East Victoria in
Australia. The town is situated 237 kilometres from
Melbourne, which is the capital city of the State of
Victoria and is very accessible by road, train, bus and
air. Although Rutherglen is home to worldrenowned forti-
ed wines such as muscats and tawny ports, its wine reper-
toire includes the full spectrum of Australian wines, and
the region is also well known for its cuisine, golf courses
and water sport activities.
The nearby towns of Albury and Wodonga within a
radius of 40 kilometres with 87 500 residents (Australian
Bureau of Statistics, 2011) further contribute to a signicant
population living in the festival's most immediate catchment
area. Positioned as the State of Victoria's original wine
festival it attracts more than 20 000 visitors and boasts
three tourism awards including Best Festival in Victoria
and was also named Best Festival in Australia (Rutherglen
Victoria, 2011). The festival, rst held in 1974, has grown
in size and stature since then and now offers an array of
entertainment and activities throughout the weekend. With
20 wineries participating, the event offers a range of wine
and other culinary experiences, product vendor booths, a
variety of entertainment activities and support services/
amenities.
A highly structured questionnaire was developed in which
the service attributes of the festival were identied based on
discussion with the event organizers and on existing service
quality literature. The questionnaire also sought information
about attendees' sociodemographics, normal wine consump-
tion behaviour and purchase of wine, food and other products
at the festival.
Information was obtained directly from the target
population of visitors (18 years and older) to the festival.
Respondents were systematically selected by approaching
every fth person who obtained a (Australian) $25 ticket
passport for wine tasting at any winery. If a visitor declined
to participate, he/she was replaced with the next fth person
and so on. Respondents were handed the questionnaire and
Wine Festival
Quality
Behaviour
(Wine Buying)
Wine Festival
Satisfaction
Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
 Festivalscape: Buying Behaviour Prediction 79

requested to complete it, an activity which took between ve
to eight minutes to complete. Research administrators were
on hand throughout to respond to any queries regarding the
questionnaire. No incentives were offered, but a written
guarantee of the condentiality of their personal information
was provided. The response rate was just below 71%. A total
of 358 respondents completed the surveys, which were
thereafter analysed using the PASW 18.0 statistical software
package.
To operationalize the research study, the following
research questions were developed:
(1) What are the differences between rsttime and repeat
visitors in terms of antecedents that could inuence their
perception of service quality, satisfaction and wine
buying at the wine festival?
(2) What are the nature and number of dimensions of perfor-
mance quality and satisfaction of the wine festival
servicescape?
(3) Is there a relationship between and inuence exerted
by the dimensions of performance quality individually
and collectively and by satisfaction within the wine
festival servicescape and the rsttime and repeat visitor
dynamic on wine sales?
RESULTS AND DISCUSSION
The results are discussed starting with the sociodemographic
characteristics of the festival visitors, followed by their buy-
ing activityrelated behaviour at the festival, and nally, ser-
vice quality perception, satisfaction and buying behaviour
are linked together after factor analysis.
Sociodemographics of festival visitors
Table 1 provides an overview of the festival visitor in terms
of sociodemographic characteristics that individually or
collectively could impact on their festival behaviour.
Table 1 indicates a predominance of female visitors (54.5%)
as opposed to male visitors (45.5%). The reasons for this were
not explored in the study, but a weighting towards female vis-
itors as wine tourists have been reported in several studies
(Bruwer and Lesschaeve, 2012 (Canada); Bruwer and Alant,
2009 (South Africa); Bruwer, 2004 (Australia); Olsen et al.,
2007 (USA)). The age distribution of the festival visitors is
slanted towards Millennials (1834 years) who account for
54%, whereas GenerationX (3544 years) account for 20%,
Baby Boomers (4565 years) for 25%, with Traditionalists
represented by only 1% of the visitors.

Table 1. Sociodemographic characteristics of the wine festival visitors

Characteristic Male (%) (n =163) Female (%) (n=195) Total (%) (n=358)
Gender: 45.5 54.5 100.0
Age group:
1824years 23.3 19.5 21.3
2528years 16.0 20.0 18.2
2934years 16.6 12.3 14.2
3539years 6.7 14.4 10.9
4044years 10.4 8.2 9.2
4554years 16.6 15.4 15.9
5565years 9.8 8.7 9.2
65+years 0.6 1.5 1.1

Education level:
No postsecondary qualication 34.6 29.6 31.8
Undergraduate postsecondary qualication 48.1 49.1 48.7
Postgraduate postsecondary qualication 17.3 21.3 19.5

Annual household income* level:

<$25000 per year 7.9 14.0 11.2
$2500049999 per year 20.0 23.9 22.1
$5000074999 per year 22.1 28.1 25.4
$7500099999 per year 17.9 15.2 16.4
$100000149999 per year 17.9 14.6 16.1
$150000200000 per year 7.1 1.2 3.9
>$200000 per year 7.1 3.0 4.9

Origin Place of Permanent Residence

Inside region (NE Victoria/Rutherglen) 32.3 28.9 30.4
Outside region (outside of NE Victoria/overseas) 67.7 71.1 69.6

Number of persons living in household: Means Means Means

Persons under 18years old 0.51 0.48 0.49
Persons 18years and older 2.21 2.11 2.15
Persons in household (total size) 2.72 2.59 2.64
Persons 18years and older who drink wine 1.84 1.86 1.85

*Australian $.

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
80 J. Bruwer

The median annual household income level in Australia is
$44 820 (Australian Bureau of Statistics, 2011), and by com-
parison, the income levels of the festival visitors are notice-
ably higher, with around 67% from households where the
annual income is in excess of $50 000. This is in all likeli-
hood largely attributable to the high level of postsecondary
education with 68% in possession of either an undergraduate
or postgraduate postsecondary qualication. The small
household size could also be a contributing factor to their
relative afuence with the mean number of persons per
household at only 2.6 (number of dependent children plus
adults), a gure which includes very few dependent children.
It is insightful to note the very high incidence of both adults
in multiperson households being wine consumers with a
mean of 1.85 persons. Finally, the origin of visitors brings
to light the fact that the overwhelming majority of 70% of
the visitors originate from outside the destination region
versus 30% of visitors from inside the North East Victoria
region. The fact that the festival attracts 68% of visitors from
other regions all over Australia provides credence to the
statement that it is a wellestablished and signicant event
on the national calendar. This is further underpinned in that
2% of the visitors are from a total of three overseas countries,
which also underlines the area's attraction for tourists.
the rst time. Contrary to what is generally expected in terms
of tourism destination choice theory when linked to distance
from origin (Nicolau and Ms, 2006), most visitors are from
outside the destination region of North East Victoria (70%),
and the highest percentage of repeat visitors (63%) is also
represented by visitors from inside the region. Chisquare
statistical testing however reveals no signicant differences
between visitors from inside and outside the region in terms
of their past attendance of the festival.
Having established that most visitors originate from
outside the destination region, the study also investigated
whether rsttime and repeat visitors differed regarding the
time frame or planning period in terms of how long before
the event the nal decision to attend was made (Table 3).
It follows that there is a highly signicant difference ( =
67.106, p = 0.000*) between the decisionmaking time
frames of rsttime and repeat visitors. For repeat visitors,
there is a signicantly longer period between making the -
nal decision and actually attending the event with 59% mak-
ing this decision at least a month before the event, whereas
36% of rsttime visitors decided within a time span of only
1 week before attending the event. In the case of repeat visi-
tors, this could be an indication of their involvement with
and likely loyalty to the festival.

Festival attendance and decision making by visitors

The prior festival attendance levels and frequency of atten- Festival buying activityrelated metrics
dance are reected in Table 2. Finally, as far as the base variables are concerned, Table 4
The attendance history of festival visitors indicates 62% exhibits the wine festival activityrelated factors for different
are repeat visitors with the balance of 38% attending for attendee groups/segments.

Table 2. Prior incidence and frequency of attendance of the festival visitors

Visitor groups by origin Test statistic

Inregion Outside region

Incidence of previous attendances Overall % (n=358) % (n=109) % (n=249) Signicance
Attended the wine festival in the past 62.0 61.0 62.7 0.016 0.900
Attended the wine festival last year 43.3 41.9 42.9 0.031 0.861
Analysis of variance

Number of previous attendances Mean Mean Mean F Signicance (2sided)

Number of times previously attended 2.88 2.59 3.03 0.724 0.395

Grouping variable: North East Victoria Regionbased versus Outside North East Victoria Regionbased visitors

Table 3. When nal decision was made to attend the festival

Visitor groups by festival attendance history Test statistic

Timing of decision Overall % (n=358) Firsttime % (n=136) Repeat % (n=222) d.f. Signicance
(6.4) (11.0) (3.6)
Today 6.4 11.0 3.6
Yesterday 7.0 (13.4) 6.6 (17.6) 7.2 (10.8)
During the past week 13.1 (26.5) 18.4 (36.0) 9.9 (20.7)
During the past two weeks 11.2 (37.7) 16.2 (52.2) 8.1 (28.8) 67.106 7 0.000*
During the past month 17.0 (54.7) 24.3 (76.5) 12.6 (41.4)
One to threemonths ago 15.6 (70.3) 17.6 (94.1) 14.4 (55.8)
Four to sixmonths ago 3.9 (74.2) 2.2 (96.3) 5.0 (60.8)
More than sixmonths ago 25.8 (100.0) 3.7 (100.0) 39.2 (100.0)

*Signicant at 0.05 level; ( ) cumulative %; Grouping variable: rsttime versus repeat visitors

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
 Festivalscape: Buying Behaviour Prediction 81

Table 4. Buying activityrelated metrics and festival attendee groups

Festival buying activityrelated factor Total (n=358) Firsttime (n=136) Repeat (n=222) Signicance (twosided)
Bought wine to take away 54.7 52.9 55.9 .289 0.591
Bought same wine in retail in last threemonths 21.4 12.5 26.6 5.389 0.020*
Fvalue Signicance (twotailed)
Total amount spent on wine ($) 86.28 56.74 103.44 7.611 0.006*
Number of bottles of wine bought 5.70 3.44 7.01 5.481 0.020*
Average price paid per bottle ($) 15.14 16.49 14.76 .321 0.571
Total amount spent at the festival 109.89 82.88 125.22 4.977 0.026*

Festival buying activityrelated factor Total (n=358) Male (n=163) Female (n=195) Signicance (twosided)
Bought wine to take away 54.7 54.0 55.4 0.070 0.791
Bought same wine in retail in last threemonths 21.4 29.5 14.8 6.249 0.012*
Fvalue Signicance (twotailed)
Total amount spent on wine ($) 86.28 85.32 87.07 0.011 0.917
Number of bottles of wine bought 5.70 5.18 6.12 0.394 0.531
Average price paid per bottle ($) 15.14 16.47 14.23 0.024 0.876
Total amount spent at the festival 109.89 103.56 115.29 0.405 0.525

Festival buying activityrelated factor Total (n=358) <35years (n=192) 35years (n=166) Signicance (twosided)
Bought wine to take away 54.7 54.7 54.8 0.001 0.980
Bought same wine in retail in last three months 21.4 17.1 26.4 2.467 0.116
Fvalue Signicance (twotailed)
Total amount spent on wine ($) 86.28 56.31 120.86 16.225 0.000*
Number of bottles of wine bought 5.70 3.47 8.27 10.972 0.001*
Average price paid per bottle ($) 15.14 16.23 14.61 0.021 0.889
Total amount spent at the festival 109.89 73.61 151.99 19.130 0.000*

Grouping variables: gender, age and rsttime versus repeat visitation;

*Signicant at 0.05 level.
All currency values in Australian $.

This analysis provides context to the core evaluation of
the study that follows, which is to determine the relationship
between the quality perception of festival elements and over-
all satisfaction and their inuence (or not) on wine buying at
the event.
From Table 4, it follows that rsttime and repeat visitors
differ signicantly as far as having bought the same wine at
the festival as in retail within a threemonth period prior to
the festival. Whereas only slightly more repeat than rsttime
visitors (56% vs 53%) bought wine to take home at the festi-
val, the difference of having an established connection with
the brand(s) between repeat (27%) and rsttime visitors
(13%) is signicant (p = 0.020). It is also insightful that re-
peat visitors bought signicantly more bottles of wine (seven
bottles) and spent signicantly more money on wine
($103.44) than rsttime visitors. Repeat visitors also spent
signicantly more money overall on everything at the festi-
val ($125.22) than rsttime visitors. These ndings clearly
underline some of the differences between rsttime and re-
peat visitors and, hence, the rationale for using them as pre-
dictor variables in the festivalscape service quality model.
Although gender is intertwined with many aspects of
human behaviour (MeyersLevy and Maheswaran, 1991),
and differences between male and female visitors have also
been proven in wine consumer behaviour (Bruwer et al.,
2011), this study did not identify many statistically signi-
cant differences among festival attendees by gender. One
exception is the signicantly higher incidence of male
(30%) versus female visitors (15%) who bought the wine in
retail during the threemonth period prior to the festival.
Marginally, more female visitors (55%) in comparison with
male visitors (54%) bought wine at the festival, and female
visitors also bought more bottles and spent more on wine
than male visitors (Table 4). Male visitors bought fewer
bottles of wine than female visitors (difference not statisti-
cally signicant), but their average amount spent per bottle
was higher than that of female visitors. Although the under-
lying reasons were not pursued in this study, it is possible
that what was at play here was a brand loyalty riskreduction
strategy versus experimenting behaviour.
When the age demographic is considered, the differ-
ences are quite profound, as has been the case in other stud-
ies (Bruwer et al., 2011). Whereas the incidence of wine
buying at the festival barely differs between the older than
35 years group (54.8%) and the younger Millennial group
(54.7%), older visitors outperform the younger visitors
on almost all other metrics. Although the difference be-
tween the two groups is not statistically signicant as far
as prior buying in retail is concerned, the connection older
visitors has with the brand is already evident in that 26% of
them bought the wines during the threemonth period prior
versus 17% among Millennials. Comparisons between the
other wine buying metrics at the festival reveal that the
older group bought signicantly more wine (8.3 bottles),
spent more money thereon ($120.86) and spent more
money overall on everything at the festival ($151.99).
Although the underlying reason could be income related,
it has, on the other hand, also been proven that wine
consumption peaks at later age (Bruwer et al., 2011), for
reasons unknown to date.

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
82 J. Bruwer

Table 5. Festivalscape elements service quality perception ratings with each of the factors extracted (see Tables 6 and 7), and it
Standard was concluded that the scale yielded sufciently reliable
Festivalscape element Mean* deviation Var results to warrant further testing using these answers.
An assessment of the suitability of the data for factor anal-
Wineries participating 4.4188 0.6369 0.406
Helpfulness of the service staff 4.2397 0.6686 0.447 ysis was the rst step in the process. First, two main issues
Access/parking arrangements 4.1804 0.7340 0.539 were considered, namely, the strength of the relationship
Ease of moving around 4.1761 0.7399 0.547 among the items and the size of the sample. Tabachnick
Rutherglen as cenrepoint 4.1363 0.7312 0.535 and Fidell (2001) indicated that a sample size of at least
Designated driver programme 4.0761 0.8325 0.693 300 is required for factor analysis, which this study exceeded
Festival's theme 4.0508 0.7636 0.583
Maps at strategic locations 4.0206 0.8366 0.700 (n = 358). The majority of items in the dataset have correla-
Music (band/s at the wineries) 4.0028 0.8840 0.782 tion coefcients in excess of 0.5 as per the directive provided
Cost aspect overall 3.9790 0.8145 0.663 by Tabachnick and Fidell (2001).
Ticketing arrangements 3.9603 0.7306 0.534 Next, the KaiserMeyerOlkin (KMO) measure of
Festival's brochure and poster 3.8288 0.8054 0.649 sampling adequacy and Bartlett's statistical test of sphericity
Seating arrangements 3.7921 0.8171 0.668
Wines served with food 3.7797 0.7211 0.520 were used to help assess the factorability of the data.
Entertainment (excluding music ) 3.6792 0.9467 0.896 Although some of the communalities after extraction are
Lunch dish (type and portion size) 3.6629 0.9532 0.908 slightly below 0.5 (see Table 6), which could suggest that a
Restroom facilities 3.6218 1.0626 1.129 sample size above 500 is required (MacCallum et al.,
*Fivepoint Likert scale: 1=very dissatised, 5=very satised.
1999), the KMO measure of sampling adequacy is 0.857,
which is comfortably above the recommendation of 0.50
(Kaiser, 1974). In fact, this value falls within the category
(0.700.90) that can be regarded as very good according
Festivalscape factor analysis and structural model to Hutcheson and Sofroniou (1999). Bartlett's test whether
As a rst step, the data obtained from the service quality mea- the correlations between the scale items are sufciently large
surement were examined (Table 5). There are few outliers, and for factor analysis to be appropriate yielded: ( (136) =
the data were found to be fairly normally distributed. 1690.972, p < 0.000) and indicated that the correlations
Table 5 exhibits the levels of perceived quality with the within the Rmatrix are sufciently different from zero to
preidentied 17 festival service elements. A vepoint warrant factor analysis. Therefore, all the evidence suggests
Likert scale was used for this purpose. The Cronbach alpha that the sample size (n = 358) is adequate and the strength
value for the scale overall of 0.847 is comfortably above of the relationships among items sufcient to yield distinct
the minimum acceptable level of 0.70. This was also the case and reliable factors from the analysis.

Table 6. Results of factor analysis for festivalscape and overall satisfaction level
Factors and Items Factor loading Eigenvalue Variance explained % Reliability coefcient
F1: Generic features and service staff 5.142 30.249 0.812
Ease of moving around 0.833
Access/parking arrangements 0.620
Ticketing arrangements 0.459
Cost aspect overall 0.659
Wineries participating 0.760
Helpfulness of service staff 0.561
F2: Entertainment and catering 1.765 10.382 0.711
Music (band/s at the wineries) 0.548
Entertainment (excluding music ) 0.728
Lunch dish (type and portion size) 0.753
Wines served with food 0.683
F3: Comfort amenities 1.232 7.247 0.707
Restroom facilities 0.464
Seating arrangements 0.488
Maps at strategic locations 0.676
Designated driver programme 0.557
F4: Festival venue and information 1.026 6.034 0.709
Rutherglen as centre point 0.414
Festival's theme 0.702
Festival's brochure and poster 0.703
Total variance % explained 53.911
S1: Satisfaction 1.246 41.539 0.759
Expectations met overall 0.617
Intention to attend again in future 0.627
Positive feeling about attending festival 0.602

*Oblimin rotation used (allow for fact that items could be correlated) with correlation coefcients set at >0.40.

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
 Festivalscape: Buying Behaviour Prediction 83

Table 7. Correlation matrices, means and standard deviations of the factor indicators
Indicator 1 2 3 4 5
Wine festival quality perception
F1: Generic festival features and service staff 1.000
F2: Entertainment and catering 0.224 1.000
F3: Comfort amenities 0.326 0.194 1.000
F4: Festival venue and information 0.235 0.243 0.102 1.000
Wine festival satisfaction
S1: Satisfaction level overall 0.257** 0.343** 0.180** 0.237** 1.000
Reliability 0.812 0.711 0.707 0.709 0.759
Mean 4.0053 4.1591 3.7811 3.8776 4.4107
Standard deviation (SD) 0.5624 0.5194 0.6388 0.5927 0.5312
**Correlation signicant at 0.01 level (twotailed).

To investigate the underlying structure of the scale items
in the questionnaire, the data collected from the respondents
were then subjected to principal component analysis as an
extraction method, utilizing oblimin as the rotation method
with Kaiser normalization.
The results in Table 6 show that four factors with eigen-
values exceeding 1.0 were identied as underlying the scale
items measuring festival quality perception. In total, these
factors accounted for 54% of the variance. The screeplot
showed a break between the fourth and fth components,
in the process also pointing to the existence of four factors.
Finally, to be condent about the number of factors
extracted, parallel analysis was conducted using the Monte
Carlo PCA technique. The parallel analysis yielded random
eigenvalues above 1 of respectively 1.3949 (0.0458 SD),
1.3105 (0.0312 SD), 1.2572 (0.0283 SD) and 1.022
(0.0290 SD) also conrming there were four components to
the factor matrix, the same result obtained in both the
screeplot and by means of oblimin rotation with Kaiser
normalization. It was therefore decided that a fourfactor
solution was indeed optimal in the circumstances. Separate
factor analysis was conducted for visitor satisfaction with
the festival (three items), and this revealed one factor.
Reliability analysis showed acceptable alpha levels of >
0.70 for each of the factors extracted and conrmed that
the scale used to measure festival quality perception and
satisfaction was indeed reliable (Table 7).
Table 7 also provides an exposition of the pattern matrix
plus information on correlations among the factors.
The latent variables (wine buying and visitor type) are
categorical in nature, and hence, multiple regression is not
suitable for the structural model. This necessitated the use
of logistic regression to assess the impact of a set of predic-
tors such as perceived quality and overall satisfaction on a
dependent variable (wine buying). Ideally, the predictor vari-
ables will be strongly related (above r = 0.9); otherwise,
multicollinearity may exist.
The results of the logit analysis are reected in Table 8.
To determine how well the regression model performs the
goodnessoft test (omnibus test of model coefcients) sup-
ports the model as being worthwhile as good t is indicated
by a signicance value less than 0.05, which is the minimum
allowed. The Hosmer and Lemeshow goodness of t test of
model coefcients supports the model as worthwhile as it
has a value greater than .05 ( = 8.958, 6 d.f., p = 0.176).
The pseudo Rsquare statistics (from the Cox and Snell
and Nagelkerke Rsquare tests) are as follows: 0.101 and
0.134 respectively, indicating that between 10% and 13%
of the variability is explained by the set of variables. The
positive predictive value of the model is 57.0%, indicating
that of the people predicted to buy wine at the festival, the
model accurately predicted 57% of them.
The variables in the equation (Table 8) reect the contri-
bution or importance of each of the predictor variables. The
values of the Wald test indicate the variables that contribute
signicantly to the predictive ability of the model. It can be
seen that only overall satisfaction features (p = 0.008*)
contribute signicantly to the predictive ability of the model.

Table 8. Variables in the logit regression equation

95% Condence interval
for Exp (B)
Standard
B error Wald d.f. Signicance Exp(B) Lower Upper
Firstrepeat visitor 0.159 0.224 0.503 1 0.478 1.173 0.755 1.821
Generic features and service staff 0.105 0.236 0.198 1 0.657 0.901 0.568 1.429
Entertainment and catering 0.275 0.294 0.871 1 0.351 0.760 0.427 1.353
Comfort amenities 0.124 0.194 0.405 1 0.524 1.132 0.773 1.656
Festival venue and information 0.212 0.231 0.842 1 0.359 0.809 0.515 1.272
Satisfaction 0.611 0.229 7.125 1 0.008* 1.843 0.255 2.886
Constant# 0.708 1.049 0.455 1 0.500 0.493
#
Wine buying at the festival;
*Signicant at 0.05 level.

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
84 J. Bruwer

F1: Generic Festival

Features& service staff

.224
.007 -.105

F2: Entertainment and

Catering
.022 -.275

.194

First-time/ .053 .124 Behaviour

F3: Comfort Amenities
Repeat Visitor (Wine Buying)

.141**
.676 -.212
.102

F4: Festival Venue

and Information
.192**
.611

.340*

S1: Wine Festival

Satisfaction

** Correlation significant at the .01 level (2-sided)

Figure 2. Interrelationships between visitor type, perceived quality, satisfaction and buying behaviour at the wine festival.

Examining of the odds ratios (Exp B) for each of the
independent variables reveals the comfort amenities among
festival quality aspects, and rsttime repeat visitor status
are the signicant predictors with values >1. The upper level
of the 95% condence interval of the odds ratios shows a nar-
row spread (1.2722.886 and indicates a 95% certainty that
the actual value of the odds ratio lies somewhere between
1.272 and 2.886.
The Bvalues in the equation show the direction of the
relationship. A negative Bvalue indicate that an increase in
the score of the independent variable will result in a decreased
probability of the case recording a score of 1, indicating wine
buying in this case. The negative values for generic features
and service staff, entertainment and catering, and festival venue
and information indicate that the more satised overall and the
higher visitors perceive those three festival dimensions, the
lower the likelihood they will buy wine at the event.
Figure 2 shows the values and directions of the various re-
lationships identied by the statistical procedures discussed.
The results obtained by the logit regression model indicate
that actual buying behaviour is inuenced by service quality per-
ception but not to the same degree by each dimension thereof.
This is akin to previous studies (Lee et al., 2008; Cole and Illum,
2006; Baker and Crompton, 2000), etc.) who conrmed it, albeit
for the behavioural intentions construct only. The rsttime/
repeat visitor dynamic inuences the buying behaviour outcome
with repeat visitors having the greatest positive inuence.
Among the quality dimensions, only the comfort amenities
affect winebuying behaviour positively, whereas generic
festival features and service staff, entertainment and catering,
and festival venue and information have a negative relationship
with wine buying. Strong positive relationships between satis-
faction and winebuying behaviour and between quality percep-
tion overall and overall satisfaction were found.
CONCLUSIONS, IMPLICATIONS AND RESEARCH
RECOMMENDATIONS
The study conrmed four festivalscape dimensions, albeit
somewhat different from those identied by Baker and
Crompton (2000), in that it identied generic festival features
and service staff, entertainment and catering, comfort ameni-
ties, and festival venue and information as the indicators of
quality perception. A strong positive relationship between
overall quality perception and overall satisfaction was found.
A new perspective was provided for the festivalscape
knowledge base in that it identied the rsttime and repeat
visitor dynamic as a predictor of winebuying behaviour at
the festival. The higher the proportion of repeat visitors, the
higher the likelihood of wine buying will be. The quality
dimensions, whether individual or overall, are not strong pre-
dictors of winebuying behaviour. In fact, the relationships
between generic festival features and service staff, entertain-
ment and catering, festival venue and information, and wine
buying behaviour, and comfort amenities, is positive but
weak. In all, there are relatively strong correlations between

Copyright 2012 John Wiley & Sons, Ltd. Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr

the rsttime/repeat visitor dynamic, overall satisfaction with
the festival and winebuying behaviour.
The research also differs from previous studies in that
contrary to the common approach of attempting to link
behavioural intentions to quality perception and satisfaction
of a festival, it provided new insights on the link with actual
(buying) behaviour and how this is moderated by the rst
time/repeat festival visitor dynamic. As such, it provided a
direction based upon which further research on these aspects
can be conducted.
As far as the managerial implications of a festival are
concerned, it is clearly of the utmost importance to achieve
a high degree of repeat attendance as this correlates with -
nancial gains in the form of selling more wine. That
depends of course on whether the objective of the event is
to sell as much wine as possible and not to mainly promote
and create awareness.
That the service quality dimensions of generic festival
features and service staff, entertainment and catering, com-
fort amenities, and festival venue and information did not
have the same positive effect on buying behaviour as overall
satisfaction, does not mean these are relatively unimportant.
The nature and type of activities offered at a festival must
be cohesive with the event theme (i.e. wine seminars with a
wine festival), continuously evaluated for signs of impact
wearout, and renewed or replaced depending on the specic
situation.
The measures of service quality and satisfaction were not
developed using formal procedures and, hence, are a limita-
tion of this study. They were also not pretested, and the
results should thus be interpreted with some caution.
The benchmarking of a festival against others of a similar
nature, e.g. food and wine, would be difcult in reality.
Events have different key features, and the location of the
wine region where such a festival is held would impact on
the atmosphere that is created at the festival, whereas the
sociodemographics of people living in the main source
area(s) will also introduce further variability.
It is nevertheless recommended that further research
studies explore the difference(s) between behavioural
intention (as in several past studies) and actual behaviour
(as in our study) and their respective relationships with
perceived quality and satisfaction of a festival. The knowl-
edge base will also benet greatly if longitudinal research
is conducted that measures the effect of festival attendance
on actual buying behaviour postevent.
REFERENCES
Alant K, Bruwer J. 2010. Winery visitation sets: intraregional
spatial movement of wine tourists in branded wine regions. In-
ternational Journal of Wine Business Research 22(2): 191210.
Alant K, Bruwer J. 2004. Wine tourism behaviour in the context of
a motivational framework for wine regions and tasting rooms.
Journal of Wine Research 15(1): 2737.
Anderson E, Fornell C, Lehmann D. 1994. Customer satisfaction,
market share, and protability: ndings from Sweden. Journal
of Marketing 58(3): 5366.
Copyright 2012 John Wiley & Sons, Ltd.
Festivalscape: Buying Behaviour Prediction 85
Australian Bureau of Statistics. 2011. Average Weekly Earnings.
Report Cat. No. 6302.0, Canberra: Australia.
Baker DA, Crompton JL. 2000. Quality, satisfaction and behavioral
intentions. Annals of Tourism Research 27(3): 785804.
Bitner MJ. 1992. Servicescapes: the impact of physical surround-
ings on customers and employees. Journal of Marketing 56(2):
5771.
Bruwer J, Lesschaeve I. 2012. Sources of information used by
tourists travelling to visit Canadian winery tasting rooms.
Tourism Planning and Development 9(3): 269289.
Bruwer J, Lesschaeve I, Campbell BL. 2012. Consumption dynam-
ics and demographics of Canadian wine consumers: retailing
insights from the tasting room channel. Journal of Retailing
and Consumer Services 19(1): 4558.
Bruwer J, Saliba A, Miller B. 2011. Consumer behaviour and
sensory preference differences: implications for wine product
marketing. Journal of Consumer Marketing 8(1): 518.
Bruwer J, Alant K. 2009. The hedonic nature of wine tourism
consumption: an experiential view. International Journal of
Wine Business Research 21(2): 235257.
Bruwer J. 2004. The love affair of generationx consumers with the
winery tasting room. The Australian and New Zealand
Grapegrower & Winemaker, 491, December, 1924.
Bruwer J. 2003. South African wine routes: some perspectives on
the wine tourism industry's structural dimensions and wine
tourism product. Tourism Management 24(4): 423435.
Bruwer J. 2002. Wine and food events: a golden opportunity to
learn more about wine consumers. Australian & New Zealand
Wine Industry Journal 17(3): 9299.
Carmichael BA. 2005. Understanding the wine tourism experience
for winery visitors in the Niagara region, Ontario. Tourism
Geographies 7(2): 185204.
Childress RD, Crompton JL. 1997. A comparison of alternative
direct and discrepancy approaches to measuring quality of per-
formance at a festival. Journal of Travel Research 36(2): 3643.
Cohen E, BenNun L. 2009. The important dimensions of wine
tourism experience from potential visitors' perception. Tourism
and Hospitality Research 9(1): 2031.
Cole ST, Chancellor HC. 2009. Examining the festival attributes
that impact visitor experience, satisfaction and revisit intention.
Journal of Vacation Marketing 15(4): 323333.
Cole ST, Illum SF. 2006. Examining the mediating role of
festival visitors' satisfaction in the relationship between ser-
vice quality and behavioral intentions. Journal of Vacation
Marketing 12(2): 160173.
Cole ST, Crompton JL. 2003. A conceptualization of the relation-
ships between service quality and visitor satisfaction, and their
links to destination selection. Leisure Studies 22(1): 6580.
Cole ST, Crompton JL, Willson VL. 2002. An empirical investiga-
tion of the relationships between service quality, satisfaction and
behavioral intentions among visitors to a wildlife refuge.
Journal of Leisure Research 34(1): 122.
Cunnell D, Prentice R. 2000. Tourists' recollections of quality in
museums: a servicescape without people? Museum Management
and Curatorship 18(4): 369390.
Derrett R. 2003. Festivals and regional destinations: how festivals
demonstrate a sense of community and place. Rural Society
Journal 13(1): 3553.
Dodd TH. 1999. Attracting repeat customers to wineries. Interna-
tional Journal of Wine Marketing, 11(2): 1827.
Getz D. 2010. The nature and scope of festival studies. Interna-
tional Journal of Event Management Research 5(1): 147.
Getz D, Cheyne J. 2002. Special event motivations and behaviour,
In The Tourism Experience, C Ryan (ed), 2ndedn Continuum:
London; 137155.
Houghton M. 2001. The propensity of wine festivals to encourage
subsequent winery visitation. International Journal of Wine
Marketing 13(3): 3241.
Hutcheson G, Sofroniou N. 1999. The multivariate social scientist,
Sage: London.
Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr
86 J. Bruwer
Jackson J, Houghton M, Russell R, Triandos P. 2005. Innovations
in measuring economic impacts of regional festivals. Journal
of Travel Research 43(4): 360367.
Jaffe E, Pasternak H. 2004. Developing wine trails as tourist attrac-
tion in Israel. International Journal of Tourism Research 6(4):
237249.
Kaiser HF. 1974. An index of factorial simplicity. Psychometrika
39(1): 3136.
Lee I, Arcodia C. 2011. The role of regional food festivals for
destination branding. International Journal of Tourism Research
13(4): 355367.
Lee YK, Lee CK, Lee SK, Babin BJ. 2008. Festivalscapes and
patrons' Emotions, satisfaction, and loyalty. Journal of Business
Research 61(1): 5664.
Lee SY, Petrick JF, Crompton JL. 2007. The roles of quality and
intermediary constructs in determining festival attendees' behav-
ioural intention. Journal of Travel Research 45(4): 402412.
MacCallum RC, Widaman KF, Zhang S, Hong S. 1999. Sample
size in factor analysis. Psychological Methods 4(1): 8499.
Macionis N. 1996. Wine tourism in Australia. In Proceedings of
Tourism Down Under II: A Tourism Research Conference.
University of Otago, Dunedin: New Zealand; 264286.
MeyersLevy J, Maheswaran D. 1991. Exploring differences in
males' and females' processing strategies. Journal of Consumer
Research 18(1): 6370.
Nicolau JL, Ms FJ. 2006. The inuence of distance and prices on
the choice of tourist destinations: the moderating role of motiva-
tions. Tourism Management 27(5): 982996.
Oliver RL. 1981. Measurement and evaluation of satisfaction
process in a retail setting. Journal of Retailing 57(3): 2548.
Olsen J, Thach L, Nowak L. 2007. Wine for my generation: explor-
ing how U.S. wine consumers are socialized to wine. Journal of
Wine Research 17(1): 118.
Petrick JF. 2004. The roles of quality, value and satisfaction in
predicting cruise passengers' behavioural intentions. Journal of
Travel Research 42(4): 397407.
Copyright 2012 John Wiley & Sons, Ltd.
Rutherglen Victoria. 2011. Rutherglen Winery Walkabout Major
Events [accessed online 13 Dec 2011 at]: http://www.
rutherglenvic.com/events/majorevent_more.asp?
majoreventID=3
Saleh F, Ryan C. 1993. Jazz and knitwear: factors that attract
tourists to festivals. Tourism Management 14(4): 289297.
Shanka T, Taylor R. 2004. Discriminating factors of rsttime
and repeat visitors to wine festivals. Current Issues in
Tourism 7(2): 134145.
Tabachnick BG, Fidell LS. 2001. Using multivariate statistics,
4thedition , HarperCollins: New York.
Telfer D. 2000. The northeast wine route. In Wine Tourism Around
the World, CM Hall, L Sharples, B Cambourne, N Macionis
(eds). Butterworth Heinemann: Oxford.
Thrane C. 2002. Jazz festival visitors and their expenditures: linking
spending patterns to musical interest. Journal of Travel
Research 40(3): 281286.
Um S, Chon K, Ro Y. 2006. Antecedents of revisit intention.
Annals of Tourism Research 33(4): 11411158.
Yoon YS, Lee JS, Lee CK. 2010. Measuring festival quality and
value affecting visitors' satisfaction and loyalty using a struc-
tural approach. International Journal of Hospitality Manage-
ment 29(2): 33542.
Yuan J, Morrison AM, Cai LA, Linton S. 2008. A model of wine
tourist behaviour: a festival approach. International Journal of
Tourism Research 10(4): 207219.
Yuan J, Jang S. 2008. The effects of quality and satisfaction
on awareness and behavioral intentions: exploring the role
of a Wine festival. Journal of Travel Research 46(1):
279288.
Yuan J, Jang S, Cai LA, Morrison AM, Linton S. 2006. Analysis of
motivational and promotional effects of a wine festival. In
Global Wine Tourism: Research, Management and Marketing,
J Carlsen, S Charters (eds). CABI: Oxon, UK: 96208.
Zeithaml VA, Berry LL, Parasuraman A. 1996. The behavioural
consequences of service quality. Journal of Marketing 60(2):
3146.
Int. J. Tourism Res. 16: 7686 (2014)
DOI: 10.1002/jtr