International Journal of Hospitality Management 43 (2014) 108–120
Contents lists available at ScienceDirect
International Journal of Hospitality Management
journal homepage: www.elsevier.com/locate/ijhosman
Wine attributes, perceived risk and online wine repurchase intention:
The cross-level interaction effects of website quality
Meehee Cho a,1 , Mark A. Bonn b,2 , Sora Kang c,∗
a
b
c
Dedman School of Hospitality, Florida State University, 288 Champions Way, UCB 4100, Tallahassee, FL 32306-2541, United States
Dedman School of Hospitality, Florida State University, 288 Champions Way, UCB 4110, Tallahassee, FL 32306-2541, United States
Department of Business Administration, Hoseo University, 78 Gackwonsa Way, Dongnam-gu, Cheonan-si, Chungnam 330-713, South Korea
a r t i c l e
i n f o
Keywords:
Wine attributes
Website quality
Perceived risk
Online wine repurchase intention
Hierarchical linear modeling
Multilevel analysis
a b s t r a c t
It is vital for the successful web-based wine business to reduce perceived risk associated with online
wine purchasing because it negatively affects repurchase intention. Randomly obtained data from 457
U.S. consumers identified as having purchased wine online was analyzed using hierarchical linear modeling (HLM) to determine the relationships between key wine attributes, perceived risk and online wine
repurchase intention. Results confirmed that sensory and origin-related attributes positively influence
perceived risk. The cross-level interaction role that wine website quality plays upon the relationship
between wine attributes and perceived risk was also examined. Results revealed that information quality and service quality moderate the impact of the origin attribute, which in turn reduces perceived risk
with online wine shopping. These findings offer useful implications for online wine website managers to
develop more effective website frameworks.
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Wine is one of the most popular alcoholic beverages in the
United States, and approximately 35% of the U.S. adult population
drink wine at an annual per capita rate of 3.03 gallons (Wine Market
Council, 2012). The U.S. wine industry has steadily increased production and sales growth annually. From 2003 to 2012, the U.S.
wine industry increased production and sales from 268.8 million
cases and $22.3 billion to 360.1 million cases and $34.6 billion, representing a 34% increase in production and a 55.2% increase in sales.
(Wine Institute, 2013). Due to this, the U.S. wine industry has been
able to build enormous business markets. Contributing to this are
the wine retailers and wineries which have expanded distribution
channels, increased market share and as a result, have established
more competitive wine industry predominance.
Given wine’s lucrative potential and growing demand in the
U.S., early online wine retailers recognized the benefits for using
e-commerce. Global customer accesses, unlimited product portfolio, low operational costs, no requirements for a physical store
∗ Corresponding author. Tel.: +82 41 560 8357; fax: +82 41 560 8308.
E-mail addresses: mcho3@fsu.edu, chom2h2@hotmail.com (M. Cho),
sorak@hoseo.edu, sorak21@gmail.com (S. Kang).
1
Tel.: +1 850 645 9338; fax: +1 850 644 5565.
2
Tel.: +1 850 567 1826; fax: +1 850 644 5565.
http://dx.doi.org/10.1016/j.ijhm.2014.09.002
0278-4319/© 2014 Elsevier Ltd. All rights reserved.
and the ability for customized marketing opportunities were compelling reasons that convinced online merchants to begin to sell
wine (Swartzberg et al., 2000). From a consumer’s perspective,
online wine shopping provides the benefit of offering a wider wine
selection because of convenient access to numerous wine websites
often driven by online merchants advertising their wine products
(Gebauer and Ginsburg, 2003). Another primary benefit for buying
wine online is that consumers do not feel pressured into making
hasty decisions. Online wine shopping allows consumers to conveniently research wines prior to placing their order (Stefalo, 2011).
When the Internet first emerged as a new source for consumer
product purchasing, an early research study by Alpin (1999) documented that the online wine business could become a powerful
channel to search and acquire desirable wine products in a relatively cost effective and convenient manner. Subsequent studies
also supported the importance online wine retailing offered as
a sales channel for becoming a successful business (Bruwer and
Wood, 2005; Stricker et al., 2001). However, since the advent of the
Internet, offline wine retail purchases have accounted for about 95%
of total annual U.S. wine sales primarily because of the U.S. Constitution’s 21st Amendment. Enacted in 1933, this federal law gave
individual states the authority to govern themselves regarding the
sale of alcoholic beverages. State laws pertaining to a three tier distribution system for the sale of alcohol products were subsequently
established. Now that U.S. consumers have embraced the Internet as a convenient channel for purchasing products, the three-tier
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
alcohol distribution system has become antiquated, highly controversial, and represents a very real threat to U.S. businesses that
sell wine online directly to consumers (Bensinger, 2012). In recognition of this situation, legislative modifications have occurred in
selected states such as California, regarding the online distribution
of wine sales that have facilitated recent sales growth for online
wine businesses (Wiseman and Ellig, 2007).
Wine.com, the top online wine retailer, is now able to ship wine
directly to consumer households living in states allowing residents
to buy wine online and then receive their online order at their residence. As a direct result of these changes, Wine.com’s online wine
sales business has more than doubled in five years from 2007 to
2012 (Burg, 2012). Following suit, Amazon.com launched online
wine marketplaces in the U.S. during October, 2012 providing a
growing awareness from a well-known Internet icon about online
wine purchasing opportunities (Pfanner, 2013).
Despite these encouraging indicators of successfulness
regarding online wine sales, several important and significant
concerns remain to be fully understood in order to further grow
U.S. online sales. The issues surrounding wine attributes and the
appropriateness of wine as an online product have been questioned
which could lead to high levels of perceived risk with online wine
purchasing (Santos and Ribeiro, 2012).
Regarding wine attributes and online purchasing behavior,
research conducted in the early stages of consumer Internet immaturity posited that wine may not be appropriate for e-commerce
sales because wine involves sensory-related attributes that must
be used in order to assist in the choice process, which is not as
important when selecting online products that are less dependent
upon sensory-related attributes (Stricker et al., 2001). Since wine
attributes depend in part, upon a subtle combination of factors
such as vintages, grape varietals, regions and wine making processes, much of these wine intangibles are unsubstantiated until
the actual bottle has been opened and the wine aroma and taste
have been evaluated (Jagle, 2012). Other wine research suggested
that because an individual consumer’s evaluation pertaining to a
certain wine may vary according to taste and preference, formulating an objective sensory description would not be possible until
the experience has been tangiblized using the senses of sight, smell
and taste (Charters and Pettigrew, 2005).
Perceived risk regarding online shopping has been associated
in earlier research with factors such as the online delivery system, privacy and security, and the inability to taste, feel, smell or
see actual products regardless of the recognized convenience and
benefits associated with online channels (Tan, 1999; Miyazaki and
Fernandez, 2001). Since wine is classified as an experiential product especially dependent upon sight, smell and taste, online wine
research regarding perceived risk has also suggested that wine
experience would be vital in making purchase decisions (Gupta
et al., 2004), and that consumers lacking experience with sensory
aspects related to wine perceive a high level of risk with online
wine purchasing (Bhatnagar and Ghose, 2004; Phau and Poon,
2000). Much more recent challenges pertaining to purchasing wine
online involve issues surrounding product authenticity. Attributes
associated with the assurance of wine’s authenticity including
guaranteed governmental classifications, varietals, labels, origin
and vintage have all become highly important in recent years due
to wine counterfeiting practices (Chow, 2013). As these wine counterfeit practices continue, today’s wine consumers may hesitate to
purchase wines using online channels.
To address this perceived risk issue, early online retailers
began focusing on establishing well-designed website frameworks
offering high levels of consumer service quality practices implemented to formulate more positive perceptions about purchasing
e-commerce products (Palmer, 2002). Numerous e- commerce
studies have emphasized the importance of website quality and
109
the role it plays in reducing perceived risk and in facilitating relationships between online consumers and companies (Cronin et al.,
2000; Lee and Kozar, 2006; Wu et al., 2003).
In the e-commerce business context, service quality can be
defined as how well the consumers’ evaluation and perception of
a website’s features reflect excellence and quality (Santos, 2003).
In contrast with traditional offline market settings where service
quality is evaluated based upon the consumers’ experience and
interactions with employees, online service quality is limited to the
consumers’ interactions with websites (Kim et al., 2006). Therefore,
evaluations of website quality can strongly depend on consumer
experiences with the online shopping process which include technical adequacy and support, privacy, visual appearance, navigation,
credibility, information quality, customization and contact interactivity (Liu and Arnett, 2000; Santos, 2003; Zeithaml et al., 2002).
Kim et al. (2008) demonstrated that system quality, information
quality, reputation, recommendation and buyer’s feedback of websites are found to be primary determinants for making purchase
decisions when shopping online.
Given the increase of online wine retailing, online wine business managers should understand those critical factors necessary
to create and maintain positive consumer impressions about the
online wine shopping experience. Thus, it is important to acquire a
more accurate understanding of the various roles overall consumer
perceptions of website quality plays upon intentions to purchase
(and repurchase) wine online. Recognizing that the importance of
consumer willingness to repurchase has been well-demonstrated
in the traditional business context, consumer repurchasing behavior has been regarded as the critical factor for businesses to obtain
a competitive advantage for retaining consumers and increasing
profitability in the online context (Hellier et al., 2003; Khalifa and
Liu, 2007; Tsai and Huang, 2007). Therefore, how perceived risk
with online wine shopping affects repurchase intention should be
examined as an important and challenging issue for online wine
business success.
Early research limited its focus to studies addressing consumer
motivation and perceptions regarding the online wine purchase
process (Gebauer and Ginsburg, 2003). Extant literature has yet
to explore the effects wine attributes have upon perceived risk
regarding online wine shopping and the degree to which website
quality affects these relationships. Therefore, this study attempts to
fill the existing void by examining website quality to determine if
improved perceptions lead to repurchase intentions. In light of this
documented evidence, the objectives of this study are to: (1) examine the effects wine attributes have upon perceived risk with online
wine shopping, (2) explore the impact perceived risk has upon
online wine repurchase intention and (3) determine moderating
effects website quality may have upon the relationships between
wine attributes and perceived risk with online wine purchasing.
This study’s sample consisted of consumers that purchased wine
using specific websites. Website quality was evaluated based upon
consumer’s prior online wine shopping experience at these specific wine websites. Because perceptions about the quality of wine
websites facilitate the formulation of a comprehensive corporate
image that could significantly affect an individual’s beliefs and
repurchase intentions, it was necessary to develop a multilevel
model designed to integrate individual-level factors involving wine
attributes, perceived risk and online wine repurchase intention
with firm-level factors involving website quality.
Therefore, a multilevel approach was deemed appropriate to
effectively investigate online wine repurchase intention (Bryk and
Raudenbush, 1992). This multilevel model was then tested to
identify relationships among individual-level variables and the
cross-level interaction mechanism between the individual and
firm-level factors (see Fig. 1). Using this method, study findings will
be able to subsequently offer actionable managerial insight to use
110
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Firm-Level
Website Quality
H3
Individual-Level
Wine Attributes
Perceived Risk
H1
H2
Online Wine
Repurchase
Intention
Fig. 1. Multilevel research model.
for developing wine website operational strategies. Results could
then lead to more frequent and enjoyable visits to wine websites
by consumers, which will ultimately enhance online wine sales.
2. Literature review and hypotheses
2.1. Wine attributes
The majority of consumer behavior studies have focused on
identifying important and different product attributes because
consumer purchase decisions are influenced by various product
characteristics (Brown et al., 2003). Previous wine researchers
have also attempted to understand factors affecting wine purchase
behaviors and have identified taste, aroma, origin, grape variety,
labeling and brand as important wine attributes influencing wine
choice behaviors (Dodd et al., 2005; Hall et al., 2001; TzimitraKalogianni et al., 1999; Quester and Smart, 1998).
Keown and Casey (1995) suggested that the key determinants
of wine choice are sensory-related attributes. In studies conducted
by Loureiro (2003) and Mueller et al. (2010), taste, aroma, flavor
and color were found to be primary components of the sensoryrelated attribute. They also emphasized the importance of wine
origin representing region, grape variety, type and vintage because
these origin factors are significantly related to wine sensory characteristics. In line with these findings, Dean (2002) confirmed that
origin is the salient attribute influencing wine purchase decisions.
In addition, an existing body of research has thoroughly addressed
the importance wine reputation plays upon perceptions about wine
quality. The reputation attribute has included price (Hall et al.,
2001), brands (Quester and Smart, 1998), wine labels (Hall et al.,
2001) and awards, ratings and evaluation scores received (Siegrist
and Cousin, 2009). These reputation-related attributes have been
found to be highly correlated with wine choice decisions because
they enable consumers to predict wine quality (Oczkowski, 2001).
2.2. Perceived risk and online wine repurchase intention
Given the Internet’s rapid growth as the newest shopping alternative, researchers have attempted to identify consumer benefits
of electronic commerce and have revealed that functional aspects
including convenience, unique merchandize offerings and great
variety of products are the primary motivational factors for online
shopping (Menon and Kahn, 2002; Wolfinbarger and Gilly, 2001).
Despite the apparent benefits of online shopping, a large number of consumers still hesitate to purchase products or service in
web-based markets because they perceive some degree of risk
(Tan, 1999; Bhatnagar and Ghose, 2004; Chang and Chen, 2008).
E-commerce marketers continue to struggle with reducing consumer perceived risk, which has been a highly prevalent academic
research topic regarding online shopping (Huang et al., 2004; Lim,
2003).
The concept of perceived risk has been historically defined with
respect to uncertainty and anxiety about the unfavorable consequences of purchasing products (Cox and Rich, 1964; Taylor, 1974).
More recently, applied to the web-based context, Forsythe and Shi
(2003, p. 869) defined perceived risk with Internet shopping as “the
subjectively determined expectation of loss by online customers
in contemplating a particular online purchase.” Extant online consumer behavior research demonstrated that consumers perceive a
higher level of risk with online shopping compared to traditional
offline shopping due to the inability to physically inspect products (Bhatnagar and Ghose, 2004; Featherman and Pavlou, 2003).
Accordingly, they concluded that higher levels of perceived risk
have more negative effects upon online purchase intention. Other
studies have empirically proven the negative relationship between
perceived risk and consumer’s online shopping intention (Lee and
Tan, 2003; Pavlou, 2003).
In line with findings reported in research literature that have
addressed perceived risk, subsequent studies have demonstrated
that perceived risk in online shopping includes multidimensional
components (Bhatnagar and Ghose, 2004; Forsythe and Shi, 2003).
Especially, with respect to dimensions of perceived risk toward
online shopping, financial risk representing the potential monetary loss is identified as the major barrier for online purchasing
(Forsythe et al., 2006). Time/convenience risk is considered to be
possibly due to the loss of time and inconvenience arising from
difficulty with searching appropriate online stores and submitting
orders, or potential delays and difficulties in delivering/receiving
products ordered on time (Forsythe and Shi, 2003). Product performance risk derives from concerns about an inappropriate product
choice because online stores do not offer consumers opportunities
to accurately evaluate the quality of products (Huang et al., 2004).
Levels of perceived risk can also be affected by product types
or their important attributes, since online shoppers must make
purchase decisions without touching, feeling, tasting, smelling or
trying products (Schiffman and Kanuk, 2006). Huang et al. (2004)
proposed that the inability to make sensory inspection leads to a
high level of perceived risk in online food or beverage purchasing.
In terms of the product of wine, perceived risk and risk-reduction
strategies have been explored in various settings including lifestyle
segments (Johnson and Bruwer, 2004), wine purchasing decisions
in restaurants (Lacey et al., 2009) and occasion-based purchasing
behavior (Bruwer et al., 2013). These studies identified different
types of perceived risk when purchasing wine such as functional
(taste), social (approval of others), financial (price) and physical
risk (hangover). Risk-reduction strategies were then suggested as
ways to reduce perceived risk when buying wine.
Therefore, it should not be surprising that perceived risk with
online wine shopping greatly influences purchase decisions. This
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
111
Table 1
Website quality models.
Models (authors)
SITEQUAL (Yoo and Donthu, 2001)
Webqual (Barnes and Vidgen, 2002)
eTailQ (Wolfinbarger & Gilly, 2003)
DeLone and McLean’s (2003) Success Model
e-S-Qual (Parasuraman et al., 2005)
NetQual (Bressolles, 2006)
E-S-QUAL (Boshoff, 2007)
Website quality factors
System quality
Information quality
Service quality
System quality
Service quality
System quality
Service quality
paper expects to find that perceived risk with online wine purchasing will fluctuate depending upon the consumer’s perceived
importance of wine attributes, which will negatively affect their
online wine repurchase intention. Based upon the above discussion,
this study suggests the following hypotheses:
H1. Wine attributes influence perceived risk with online wine
purchasing
H1a. The sensory-related attribute influences perceived risk with
online wine purchasing
H2b. The origin attribute influences perceived risk with online
wine purchasing
H3c. The reputation attribute influences perceived risk with online
wine purchasing
H2. Perceived risk negatively influences online wine repurchase
intention
2.3. Website quality
Website quality is associated with how consumers evaluate
whether a website meets their needs and reflects the overall purchase performance (Ahn et al., 2007; Chang and Chen, 2008).
First-generation website evaluation research emphasized the product and/or system aspects of customer service by introducing
models such as WebQual (Barnes and Vidgen, 2002) and SITEQUAL (Yoo and Donthu, 2001). Given the importance of consumer
participation regarding self-expression and interaction with websites, second-generation website quality research has devoted
much more attention to e-commerce consumer experiences by
encompassing a broader range of website quality scales such as
eTailQ (Wolfinbarger and Gilly, 2003), the D&M IS Success Model
(DeLone and McLean, 2003), e-S-Qual (Parasuraman et al., 2005)
and NetQual (Bressolles, 2006). Table 1 provides a summary of
website quality research models.
In particular, the D&M IS Success Model (DeLone and McLean,
2003) was designed to incorporate consumer service aspects that
included individualization, completeness, appropriateness, and
ease of comprehension with system quality and information quality. In support of DeLone and McLean’s comprehensive conceptual
model, numerous subsequent empirical website quality studies
demonstrated that website quality is a multi-dimensional construct comprised of different features representing system quality,
information quality and service quality (Bernroider, 2008; Kim and
Stoel, 2004; Lin, 2008; Tsai et al., 2011; Xu et al., 2013; Wang, 2008).
System quality and information quality relate to usability, accurate information and transaction security (Palmer, 2002; Park and
Kim, 2003). Service quality relates to customized marketing services that enable websites to attract and retain consumers (Long
and McMellon, 2004; Verhoef and Langerak, 2001). Subsequent
Ease of use, esthetic design, processing speed, security
Usability, design, information, trust
Web site design, reliability/fulfillment, privacy/security, customer service
Accessibility, availability, reliability
Completeness, ease of understanding, relevance
Assurance, customization, responsiveness
Efficiency, system availability
Fulfillment, privacy
Information, ease of use, design, security/privacy, interactivity/personalization
Efficiency, speed, system availability
Reliability, delivery, privacy
studies have consistently adopted this three dimensional construct
to judge website quality (Ahn et al., 2007; Udo et al., 2010).
First, system quality is described as a user-friendly system in
online shopping and evaluated by consumer perceptions about
access convenience, ease of use and reliability (Bharati and
Chaudhury, 2004). User-friendly system specifications can assist to
establish a positive interaction between online retailers and their
consumers (Lin, 2007; Palmer, 2002). Because of this, Ahn et al.
(2007) suggested that system quality positively influences consumer trust in online stores which in turn reduces perceived risk
with online shopping.
Second, information quality is defined as the degree of specific details about products or service offered by websites (Chiu
et al., 2005; Liu and Arnett, 2000). Studies conducted by Li et al.
(2002) and Park and Kim (2003) suggested that websites should
provide accurate, clear, relevant, timely and up-to-date information for online consumers to compare products and further enhance
consumer’s positive online purchase intention.
Third, service quality is evaluated by overall consumer perceptions about personalized and customized services presented in the
overall online purchasing processes (Ahn et al., 2007). DeLone and
McLean (2003) stated that service quality should be developed
based upon the customer-centered marketing perspective.
2.3.1. Cross-level interaction effects of website quality
In a virtual e-commerce environment, websites are the first
and primary contact tool for interaction with potential consumers.
Websites have become a sole means for delivering information,
communicating with potential consumers and increasing product
awareness (Ness, 2006). Because a website’s system that interfaces
with its consumers can play critical roles in reducing perceived risk
with online shopping (Brown et al., 2003), many US online companies have made efforts to develop effective websites which can
build positive relationships with consumers and further enhance
online sales revenue (Nowak and Newton, 2008).
Stefalo (2011) documented that consumers have difficulty with
choosing wine due to information pertaining to taste descriptions,
brands, regions and grape varietals, which can be voluminous. Furthermore, certain varietals share many similar wine attributes.
Therefore, in order to choose exactly the right wine for the right
occasion, consumers tend to search for wine-related characteristics using available information presented on websites. Due to
the inability to taste wine, consumers tend to seek websites that
provide intensive and high quality information to assist them in
their favorite wine choices. Also, they are more likely to visit websites offering advice or recommendations to assist them in reducing
perceived risk in order to simplify the wine choice decision process (Siegrist and Cousin, 2009). Therefore, information quality and
service quality with respect to how it is presented on a website
become critical factors in the success of online wine businesses.
Additionally, Mueller and Szolnoki (2010) noted that consumers’
112
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
perceived risk with buying wine online is impacted by complicated
purchase processes, undesired delivery systems (i.e., difficulties
in returning and necessary postponement of consumption), and
privacy issues. Therefore, system quality representing reliability,
adaptability and ease of use should be essential factors for improving consumer’s perceptions about online wine shopping.
Lim (2003) argued that well-established website quality can
improve favorable impressions and beliefs about online retail products, which in turn reduces the consumer’s perceived risk. Albert
(2004) empirically demonstrated that levels of perceived risk in
product performance were reduced when consumers perceived
high levels of website quality. With the increasing impacts of
e-service quality on consumer beliefs and satisfaction, the importance of website quality has been identified in e-commerce
research as essential for reducing levels of perceived risk (Cenfetelli
et al., 2008; Kettinger et al., 2009). Further research demonstrated
that website quality is a significant determinant of perceptions
about online product quality (Nowak and Newton, 2008). Because
of this, system quality, information quality and service quality provided by online wine sites may influence perceptions about wine
quality. Therefore, this study expects that perceived website quality moderates the impacts wine attributes play upon perceived risk
with online wine purchasing. Based upon this rationale, we propose
the following hypotheses.
H3. Website quality negatively moderates the impacts of wine
attributes on perceived risk with online wine purchasing such
that the relationship becomes weaker with increasing levels of
perceived website quality
H3a. System quality negatively moderates the impacts of wine
attributes on perceived risk with online wine purchasing
H3b. Information quality negatively moderates the impacts of
wine attributes on perceived risk with online wine purchasing
H3c. Service quality negatively moderates the impacts of wine
attributes on perceived risk with online wine purchasing
3. Method
3.1. Data collection and sampling
The hypotheses were tested using an online cross-sectional survey including measures for all of the variables. This study based
its sampling methods upon previous online survey research which
restricted participation to only invited subjects who were online
wine purchasers and who were also voluntarily willing to complete the survey questions (Liang and Lim, 2011; Chiou et al.,
2005). The survey subjects were randomly intercepted by student
administrators assigned to randomly selected off-premise retail
establishments that sold alcoholic beverages. These businesses
were all located in a major metropolitan city in the southeastern
U.S. During a three week period, random days, sites and times, were
used to intercept consumers whom were departing retail establishments that sold alcohol beverages, including wine. All randomly
intercepted individuals were informed of the intercept purpose and
that the process would require less than two minutes of their time,
with no incentive provided.
Consumers indicating their willing to participate were screened
using a quota sampling method consisting of age categories (Hair
et al., 2000). Additional screening questions were used in part, to
validate that consumers had purchased wine online over the previous 12 months, were able to recall and provide the website name
used for their most recent online wine purchase and were willing
to provide their email addresses for the purpose of being contacted
in the near future to participate in an online survey regarding
online wine purchasing behavior. Once all screening information
was gathered from potential purchasers of online wine, their email
addresses were entered into a data base from which invitation
emails were then sent to each of these potential participants. All
individuals were provided information pertaining to the study’s
purpose, its average time to complete and the link to the survey
which included an introductory page, a privacy policy statement
and instructions for completing the survey.
In order to avoid the possibility of multiple responses, all
study respondents were validated by matching their URL address
with information they previously provided during the screening
process, which included home zip codes (Andrews et al., 2003;
Couper, 2000). URL addresses were then used to eliminate multiple responses from the same location. In addition, the engagement
of the participants was assessed by an instructional manipulation check (IMC) to ensure participants were thoroughly reading
instructions (Oppenheimer et al., 2009). Participants were red
flagged and eliminated if they answered the IMC rather than
responding as instructed. A total of 21 respondents were removed
from the final analysis due to failing this test. The main online
survey was conducted during a two-week period.
Respondents were asked to provide names of online wine sites
where they most recently purchased wine. Respondents were then
grouped based upon those specific online wine sites they reported
in their survey information. A total of 463 usable responses were
obtained by respondents indicating having had experience buying
wine on 33 online wine sites located in the United States representing online wine shops (28.5%), online wine clubs (28.5%) and winery
websites (43%). The overall sample generated a data set where it
was found that five online wine sites contained less than four individuals. Because these five particular wine sites did not meet the
general conditions and recommendations for having a minimum of
four observations per group (Hofmann et al., 2003; Hox, 1998), they
were excluded from the sample’s data set. A total of 457 individuals
representing 28 online wine sites were thus used for hypotheses
testing.
The average sample size per group was 15 individuals and
ranged in size from 5 to 49 respondents. Demographic characteristics of the overall respondents represented females (50.5%)
having earned an undergraduate or post-graduate degree (92.2%).
Age groups were reported to include individuals between 31 and
40 years of age (26.9%), followed by respondents classified as being
41–50 years old (23.6%), 51–60 years old (21.9%), 21–30 years old
(13.6%) and 61 years old or older (14.0%). About 65% of all respondents spent below $50 when purchasing wine online by the bottle,
while 11.8% of these indicated spending over $100 per each online
wine purchase. Respondents indicated that they purchase between
1 and 3 bottles of wine (87.3%) about 1–2 times per month (91.7%)
via online wine sites. Respondents indicated using the Internet 4.75
hours per day, on average. Comparing these findings to prior online
wine buying research, many similarities existed (Bruwer and Wood,
2005; Santos and Ribeiro, 2012; Vizu, 2007). Thus, it was cautiously
concluded that this study’s sample data is representative of online
wine buying consumers (see Table 2).
3.2. Measures
A hierarchical linear model was designed to integrate both individual and firm-level variables which were developed to meet
this study’s objectives. Hierarchical linear modeling (HLM) was
specifically used because of its ability to simultaneously examine relationships within and between groups using hierarchically
nested data, while efficiently considering variance differences at
multi-levels (Woltman et al., 2012). This study’s sample included
individuals having experience in online wine shopping and represented consumers buying wine on specific websites. Likewise,
individuals in our sample data were not randomly assigned to
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Table 2
Socio-demographic characteristics and online wine purchasing behavior.
Numbers
Percentages
Socio-demographic variables
Gender
Female
Male
231
226
50.5
49.5
Age
21–30 years
31–40 years
41–50 years
51–60 years
61 years≤
62
123
108
100
64
13.6
26.9
23.6
21.9
14.0
Annual household income
$50,000≥
$50,000–$74,999
$75,000–$99,999
$100,000≤
101
103
95
158
22.1
22.5
20.8
34.6
Educational level
High school graduate
Some college/College degree
Graduate studies/Graduate degree
31
235
191
6.8
51.4
41.8
Employment status
Retired
Unemployed
Employed
64
47
346
14.0
10.3
75.7
Monthly online wine purchasing behavior
Purchases frequency
419
1–2 times
25
3–4 times
11
5–6 times
2
7–8 times
9 times≤
0
91.7
5.5
2.4
0.4
0.0
Quantity purchased
1–3 bottles
4–6 bottles
7–9 bottles
10 bottles≤
296
143
8
10
87.3
8.7
1.8
2.2
Amount spent
$50≥
$51–$100
$101–$150
$151≤
209
194
25
29
64.7
23.4
5.5
6.4
specific wine websites, but rather were assigned to wine websites based upon online stores where they previously purchased
wine. Thus, individuals within a particular wine website had the
similar buying experience of being in the same web-based environment, which may lead to increase their homogenous perceptions
and attitudes about the website. Therefore, the multilevel model
was deemed as an appropriate approach to be used for this study
(Raudenbush, 1998).
The survey instrument consisted of five parts. The first three
parts were developed to measure individual-level variables containing three constructs pertaining to (1) the importance of wine
attributes influencing wine choice, (2) perceived risk with online
wine purchasing and (3) online wine repurchase intention. The
fourth part measured firm-level variables representing perceived
website quality about online wine stores. The fifth part of this study
contained items related to socio-demographics and behaviors associated with online wine purchasing and Internet usage.
3.2.1. Individual-level variables
The importance of wine attributes was measured using 12 items
affecting wine purchasing and included taste, aroma, color, quality, region, grape varietal, wine type, brand, aging, price, label and
medal/award/rating score based upon wine attributes used in previous studies (Jaeger et al., 2010; Tzimitra-Kalogianni et al., 1999;
113
Quester and Smart, 1996). Overall perceived risk was measured
using five items representing consumer concerns about financial
loss, time and convenience loss and wine quality when consumers
buy wine online based upon studies by Wolfinbarger and Gilly
(2001) and Pires et al. (2004). An example statement was: “If I
buy wine online, I am concerned that I will not get my money’s
worth.” In terms of measuring online wine repurchase intention,
five items were adapted from Baker et al.’s (2002) study and an
example statement was “I intent to keep purchasing wine online”.
3.2.2. Firm-level variables
Perceived website quality about the online wine store was
evaluated using nine items including system quality, information quality and service quality based upon DeLone and McLean
(2003) and Bharati and Chaudhury (2004) studies. The system quality dimension consisted of items representing reliability, ease of
use and access convenience. The information quality dimension
included usefulness, timeliness and completeness related to perceptions about wine information. The service quality dimension
consisted of items related to websites providing customized services, wine recommendations and personal attention.
A total of 31 items were used in the first four parts of the study
and were all measured using a five point scale asking respondents
to indicate their level of agreement or disagreement with 1 equaling ‘strongly disagree’ and 5 equaling ‘strongly agree’. The fifth part
of this study contained socio-demographics (gender, marital status,
level of education, annual income and age), prior experience with
online wine purchasing (purpose, frequency and average amount
spent per bottle) and behavior related to Internet usage. Because
prior online shopping experience and familiarity with the Internet
has been found to have significant influences on risk and online
purchase intention (Miyazaki and Fernandez, 2001; Park and Jun,
2003), this study used frequency of online wine shopping and average Internet usage as control variables.
3.3. Data analysis: multilevel approach
HLM analysis procedures consisted of four steps based upon
prior research (Kang et al., 2012; Lyons et al., 2001). The first step
evaluated the existence of within- and between-group variance
in perceived risk with online wine purchasing by testing the null
model having no predictors (Whitener, 2001). A chi-square test on
the residual variance was performed to investigate whether the
between-group variance is significantly different than zero.
The second step tested the random intercepts and slopes of
the regression equation to explore the relationships between
individual-level variables. T-tests of parameters in level-1 models’ equations were performed to investigate relationships between
wine attributes and perceived risk and to examine the relationship between perceived risk and online wine repurchase intention.
A chi-square test on the residual variance in level-2 models was
used to confirm whether between-group variances are significantly
different from zero. In addition, this study calculated the variance
(R2 ) explained by individual-level predictors in the outcome variable to estimate a measure of effect size. For the individual-level,
group-mean centered variables were added.
The third step tested the “intercepts-as-outcomes” model was
tested to investigate the relationships between the firm-level predictor and the outcome variable. Each dimension of website quality
grand-mean centered was added to the intercept equations. T-tests
on parameters of website quality were used to confirm whether
website quality has a significant influence on perceived risk after
controlling the effects of wine attributes. Chi-square tests on the
residual variance were performed to investigate the significant
group-level variance in intercepts.
114
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Table 3
Validity and reliability results.
Cronbach’s ␣
Individual-level
Wine attributes
Sensory
Taste
Aroma
Color
Origin
Region
Grape Varietal
Type
Reputation
Brand name
Medal/Award/Rating information
Wine label
Perceived risk
If I buy wine through the internet, I am concerned that ∼
Wine will not be as beneficial as I expect
How really reliable the product will be
I will not get my money’s worth
Whether it will have the same quality as store bought wine
Purchasing wine online takes up too much of my time
Online wine repurchase intention
I intend to keep purchasing wine online
If I have an opportunity, I will repurchase wine online
I have an intention to repurchase wine online
It is desirable to repurchase wine online
Online stores are an acceptable place to repurchase wine
2 /df = 2.873 (p < .001), CFI = .942, GFI = .913, RMR = .056, RMSEA = .064
Firm-level (Website quality)
System quality
Online wine stores are always convenient to access
Online wine stores are easy to use
Online wine stores have a reliable system
Information quality
Online wine stores provide in-depth information
Online wine stores provide up-to-date information
Online wine stores provide useful information
Service quality
Online wine stores provide customized services
Online wine stores provide wine recommendations based on consumers’ preference
Online wine stores provide follow-up emails to consumers
2 /df = 1.415 (p > .05), CFI = .994, GFI = .984, RMR = .013, RMSEA = .030
Factor loadings
.734
.753
.818
.680
CCR
AVE
.818
.772
.714
.702
.808
.822
.
.685
.745
.600
.921
.700
.922
.704
.866
.685
.862
.676
.903
.758
.585
.504
.736
.721
.712
.930
.888
.885
.883
.836
.813
.864
.838
.807
.806
.777
.743
.718
.794
.749
.634
.809
.814
.801
.727
.777
.819
.794
.730
CCR, composite construct reliability; AVE, average variance extracted.
In the final step, the “slopes-as-outcomes” model was estimated to investigate the cross-level interaction effect. In this mode,
website quality was added as a predictor to slopes in the level-2
equations. T-tests were used to determine whether website quality significantly moderates impacts wine attributes have upon
perceived risk.
4. Results
4.1. Validity and reliability of measurements
First, an exploratory factor analysis (EFA) using principal component analysis with Varimax rotation was performed. Regarding
wine attributes items, one item (aging) having a lower factor loading than 0.5 along with two other items (quality and price) that
highly cross-loaded on other factors were removed. The remaining
28 items were factor analyzed again. These processes successfully
generated five factors for the individual-level and three factors for
the firm- level having eigenvalues greater than 1.
Second, a confirmatory factor analysis (CFA) was used to verify
internal and external consistency of constructs. Separate confirmatory factor analyses were performed for each of individual-level
and firm-level variables. Internal consistency of each construct estimated with coefficients of Cronbach’s alpha was ranged from 0.68
and 0.93 which surpassed the criteria for reliability acceptability
(Nunnally, 1978). As the results of the CFA presented in Table 3,
confirmatory measurement models demonstrated a good fitness.
For both the individual- and firm-level, all of CFI and GFI values exceeded the recommended 0.90. Also, values of RMSEA
were lower than the suggested 0.08 threshold. All factor loadings (>0.60), composite construct reliability (>0.70) and average
variance extracted (>0.50) were considered acceptable satisfying the recommended standards (Anderson and Gerbing, 1988;
Fornell and Larcker, 1981; Hair et al., 1998). Therefore, convergent validity and reliability of constructs used in this study were
accepted.
Table 4 presented the means, standard deviations and correlation coefficients of the eight factors including individual-level and
firm- level variables. As the result of the correlation analysis exhibited, “sensory-related”, “origin” and “reputation” attributes have
positive relationships with “perceived risk” with online wine purchasing. Perceived risk was negatively associated with values of
“online wine repurchase intention”. All three quality dimensions
about online wine sites (information, system and service quality) were negatively related to perceived risk, but were positively
associated with online wine repurchase intention. Results of the
correlation analysis supported the discriminant validity because
all values of the square root of the average variance extracted
(AVE) from all constructs were greater than correlations among
constructs (Fornell and Larcker, 1981).
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
115
Table 4
Correlations and discriminant validity.
1. Sensory
2. Origin
3. Reputation
4. Perceived risk
5. Repurchase intention
6. System quality
7. Information quality
8. Service quality
a
b
*
**
Mean
Std. dev.
1
2
3
4
5
6
7
8
4.23
3.20
3.59
2.13
3.56
3.64
3.76
3.65
.67
.55
.76
.84
.74
.58
.63
.65
.775a
.108* ,b
.189**
.359**
−.292**
−.248**
−.295**
−.269**
.765
.220**
.148*
−.151**
−.154**
−.229**
−.194**
.703
.052
.084
−.085
−.174**
−.155**
.837
−.347**
−.314**
−.569**
−.363**
.839
.402**
.437**
.457**
.828
.460**
.429**
.822
.504**
.871
Diagonal elements (in bold) are the square root of the average variance extracted (AVE).
Off-diagonal elements are the correlations among constructs.
p < .05.
p < .01.
4.2. Multilevel validity testing
To justify an appropriateness of multilevel modeling, one-way
ANOVA tests and two interclass correlation coefficients (ICC[1]
and ICC[2]) were used as multilevel validity measures. One-way
ANOVA results for each of the three firm-level variables (system,
information and service quality) indicated that the variance shared
between-group members was significantly greater than the withingroup members (F = 4.034, p < .000; F = 4.535, p < .000; F = 4.610,
p < .000). Regarding the ICCs of system quality, the ICC[1] was 0.16
and the ICC[2] was 0.75. For information quality, the ICC[1] was
0.17 and the ICC[2] was 0.78, and For service quality, the ICC[1]
was 0.18 and the ICC[2] was 0.79. Both of the ICC values for all
three website quality dimensions achieved desirable levels as suggested by previous HLM studies (Glick, 1985; Klein and Kozlowski,
2000). Results implied that the aggregation of individual perceptions about the quality of each website quality to the firm- level is
appropriate (Bliese, 1998). Therefore, this study tested the crosslevel interaction effect of all three website quality dimensions as
the firm-level predictor.
4.3. Null models testing
The preliminary step for null models testing should assure that
there are significant variances in dependent variables to analyze
cross-level effects (Woltman et al., 2012). Thus, null hierarchical
models having no level-1 or level-2 predictors were tested in order
to examine whether significant between-firms differences exist in
perceived risk and online wine repurchase intention (Hon, 2011).
Two separate null hierarchical models including perceived risk and
online wine repurchase intention were analyzed respectively as the
level-1 dependent variable. Results showed that mean values were
2.13 for perceived risk ( 00 , t = 27.93, p < 0.001) and 3.56 for online
wine repurchase intention ( 00 , t = 44.51, p < 0.001). Chi-square
tests supported that between-firm variances were significantly
different in perceived risk (2 = 76.30, p < 0.001) and online wine
repurchase intention (2 = 90.57, p < 0.001), which justified further
cross-level analyses to test our hypotheses (Luke, 2004). In addition, the ICCs suggested that the firm-level variables accounted for
12.0% of the variance in perceived risk and 20.8% of the variance in
online wine repurchase intention in this sample.
4.4. Tests for hypotheses 1 and 2
Hypothesis 1 predicted that wine attributes affect perceived
risk with online wine purchasing. First, relationships between
three wine attributes (sensory-related, origin and reputation) and
perceived risk with no firm-level variable were tested. As seen in
Table 5, results of model 1 showed that “sensory-related” and “origin” attributes have significant influences on perceived risk. The
“sensory-related” attribute ( = 0.41, p < 0.01) influences perceived
risk the most followed by the “origin” attribute ( = 0.36, p < 0.05).
However, the “reputation” attribute ( = 0.02, p > 0.1) did not have
a significant influence on perceived risk. In other words, consumers
reporting high importance about “sensory-related” and “origin”
attributes perceive high risk with online wine shopping. Therefore,
hypotheses 1a and 1b were supported, but hypothesis 1c was not
supported.
Hypothesis 2 expected that perceived risk is negatively related
to online wine repurchase intention. The relationship between
perceived risk and online wine repurchase intention at the
individual-level was tested. The result revealed that perceived
risk has a negative influence on online wine repurchase intention
( = −0.32, p < 0.01), confirming the traditional wisdom that consumers having a high level of perceived risk with online shopping
are less likely to buy wine online. Thus, hypothesis 2 was supported.
4.5. Tests of hypotheses 3
In order to test the subsequent hypotheses, the precondition
demonstrating a significant systematic between-firm variance in
the dependent variable should be met. The random-intercepts hierarchical model with three wine attributes as level-1 predictors and
with perceived risk as a level-1 dependent variable provided evidence of a significant between-firm variance (2 = 52.92, p < 0.001).
Further, three separate “intercepts-as-outcomes” models with each
dimension of website quality (as the firm-level variable) and with
three wine attributes and perceived risk (as individual-level variables) were estimated. As shown in model 2, 3 and 4 in Table 5,
results indicated that system quality ( 01 = −0.40, p < 0.1), information quality ( 01 = −0.73, p < 0.01) and service quality ( 01 = −0.51,
p < 0.01) have significant cross-level effects on perceived risk.
In addition, the slope of the “origin” attribute showed a significant between-firms variance in perceived risk (2 = 23.04,
p < 0.05). However, there was no significant variation in the
slopes of “sensory-related” (2 = 18.45, p > 0.05) and “reputation”
(2 = 14.79, p > 0.05) attributes. These results implied that the
effects “sensory-related” and “reputation” attributes have upon
perceived risk were not greatly different among this study’s sample websites. Thus, further analyses for a cross-level interaction
effect were tested on only the “origin-perceived risk” slope with
firm-level moderating variables (information, system and service
quality).
Hypothesis 3 predicted that website quality (at the firm- level)
negatively moderates the impacts wine attributes have upon
perceived risk (at the individual-level) such that the relationship becomes weaker with increasing levels of perceived quality
about online wine sites. Three separate “slopes-as-outcomes” models with each quality dimension as a level-2 moderator were
evaluated. Results of models 5, 6 and 7 (see Table 5) showed
that information quality ( 11 = −0.66, p < 0.01) and service quality ( 11 = −0.49, p < 0.05) have significant cross-level interaction
116
Table 5
Hierarchical linear modeling results.
Perceived risk
Repurchase intention
Model 1
Coeff
2.10
0.41
0.36
0.02
−0.01
−0.03
Firm level (Level-2)
SysQc ( 01 )
IQd ( 01 )
SQe ( 01 )
t
26.61**
5.48**
3.13*
0.36
−0.08
−0.99
Model 3
Coeff
t
26.80**
5.79**
3.12*
0.48
2.08
0.43
0.36
0.03
−0.03
−0.03
−0.39
−0.89
−0.40
−1.86†
Coeff
2.05
0.45
0.34
0.06
Model 4
t
37.05**
6.71**
2.85*
1.00
−0.05
−0.03
−0.68
−0.77
−0.73
−6.90**
Coeff
2.08
0.44
0.35
0.05
−0.02
−0.03
−0.51
Model 5
t
31.70**
6.02**
3.03*
0.69
−0.32
−0.39
a
†
2.08
0.44
0.33
0.03
t
26.80**
5.84**
2.86*
0.45
−0.03
−0.03
−0.42
−0.88
−0.42
−1.88†
Coeff
2.05
0.44
0.27
0.05
−0.29
0.222
1048.84
Coeff
37.08**
6.51**
2.61*
−0.78
−0.69
−0.87
−0.73
−6.82**
2.08
0.44
0.31
0.03
Model 8
t
Coeff
31.76**
6.04**
2.98*
0.46
−0.02
−0.03
−0.30
−0.94
−0.52
−3.82**
−0.49
−2.80*
t
3.64
43.47**
−0.32
0.20
0.05
−4.69**
3.08**
0.18
−0.86
0.223
1046.18
0.222
1043.74
0.226
1034.61
t
−0.05
−0.03
−0.66
0.216
1049.69
Model 7
−3.78**
Moderating effects
Origin × SysQ ( 11 )
Origin × IQ ( 11 )
Origin × SQ ( 11 )
Total R2
Deviance
Coeff
Model 6
0.229
1030.58
−3.40**
0.225
1040.42
0.157
949.89
Frequency = frequency of online wine shopping, b Internet = usage of Internet, c SysQ = system quality, d IQ = information quality, e SQ = service quality.
p < 0.1, * p < 0.05, ** p < 0.01.
Model 1
Level-1
Model 2
Level-1
Model 3
Level-1
Model 4
Level-1
Model 5
Level-1
Model 6
Level-1
Model 7
Level-1
Model 8
Level-1
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
PRij = ˇ0j + ˇ1j *(Senij ) +
ˇ2j *(Oriij ) + ˇ3j *(Repij ) +
ˇ4j *(Freij ) + ˇ5j *(Usaij ) + rij
ITij = ˇ0j + ˇ1j *(PRij ) +
ˇ2j *(Freij ) + ˇ3j *(Usaij ) + rij
Level-2
Level-2
Level-2
Level-2
Level-2
Level-2
Level-2
Level-2
ˇ0j = 00 + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(SysQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(IQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(SQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(SysQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + 21 *(SysQj ) + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(IQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + 21 *(IQj ) + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + 01 *(SQj ) + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + 21 *(SQj ) + u2j
ˇ3j = 30 + u3j
ˇ4j = 40 + u4j
ˇ5j = 50 + u5j
ˇ0j = 00 + u0j
ˇ1j = 10 + u1j
ˇ2j = 20 + u2j
ˇ3j = 30 + u3j
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Individual level (Level-1)
Intercept ( 00 )
Sensory ( 10 )
Origin ( 20 )
Reputation ( 30 )
Perceived risk
Frequencya ( 40 )
Internetb ( 50 )
Model 2
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
117
Table 6
Results of hypotheses tests.
Hypothesis
H1: Wine choice attributes → Perceived risk
H1a: Sensory → Perceived risk
H1b: Origin → Perceived risk
H1c: Reputation → Perceived risk
H2: Perceived risk → Online wine repurchase intention
H3: Website quality → Wine attributes–Perceived risk
H3a: System quality → Origin–Perceived risk
H3b: Information quality → Origin–Perceived risk
H3c: Service quality → Origin–Perceived risk
*
**
Standardized coefficient
0.41**
0.36*
0.02
−0.32**
−0.29
−0.66**
−0.49*
Result
Partially supported
Supported
Supported
Not supported
Supported
Partially supported
Not supported
Supported
Supported
p < 0.05.
p < 0.01.
effects on the relationship between the “origin” attribute and
perceived risk. However, the effect of system quality was not significant ( 11 = −0.29, p > 0.1). Therefore, hypotheses 3b and 3c were
supported, but hypothesis 3a was not supported (Table 6).
5. Conclusions and discussion
This study’s findings further contribute to the existing wine
consumer behavior literature by providing empirical support for
the role that value of website quality plays when applied to the
web-based wine business context. Although perceived risk pertaining to online wine shopping has been previously examined as a
research topic, existing studies have neglected to investigate the
relationship between wine attributes, perceived risk and online
wine repurchase intention. This study was the first attempt to do
so. In addition, this study provides a more heuristic understanding
pertaining to the moderating role that wine-related website quality
has upon the relationship between wine attributes and perceived
risk. Specifically, rather than focusing only upon consumer perceptions at the individual-level, this study employed a multilevel
approach integrating both the individual and the firm-level perspective.
Regarding the individual-level analysis, two hypotheses were
tested and results suggested wine attributes influence perceived
risk, which in turn influences online wine repurchase intention.
Another hypothesis was tested that proposed website quality at the
firm-level would have a significant cross-interaction effect on the
relationship between wine attributes and perceived risk. Overall,
results provide support for using a multilevel hypothesis testing
approach.
5.1. Theoretical implications
Results indicate that individuals placing high importance
on “sensory-related” and “origin” wine attributes would likely
perceive higher levels of risk with online wine shopping. Sensoryrelated attributes were found to have the strongest influence on
perceived risk supporting previous study findings (Stricker et al.,
2001; Huang et al., 2004). Despite our anticipated finding that
“sensory-related” and “origin” attributes both had significant influences on perceived risk, this present study discovered that the
“reputation” attribute had no significant influence upon perceived
risk pertaining to online wine shopping. This could imply that wine
brands, labels, awards, ratings and evaluation scores may not affect
perceived risk. Perhaps this could indicate that certain consumers
may already understand what it is they seek online suggesting that
an individual’s prior experience with or knowledge about wine may
be important factors in determining repurchase intention. Future
research needs to address the roles wine knowledge and prior experience with wine plays upon online wine repurchase intention.
Second, results confirmed Hypothesis 2 which predicted that
a higher perceived risk results in a lower online wine repurchase intention. Behavioral intentions are strongly related to actual
behavioral consequences (Ajzen and Driver, 1992; Vitell et al.,
2001). The impact of perceived risk on purchase intention to predict actual purchase behaviors and to investigate purchase patterns
has been demonstrated in a variety of e-commerce settings (Bai
et al., 2008; Lim, 2003). Because research has documented that
reducing consumer perceptions about risk is crucial for successful web-based businesses (Kim et al., 2008), online wine website
managers should attempt to examine opportunities to decrease
consumer’s perceived risk pertaining to online wine shopping.
Third, the cross-level interaction effects of website information,
service and system quality on the relationship between the “origin” attribute and perceived risk were tested. Results indicated
that the relationship between the “origin” attribute and perceived
risk was affected by perceptions about online wine site quality of
information and service. However, even if system quality has a
significant direct effect to reduce perceived risk, system quality
failed to demonstrate a significant cross-level interaction effect.
When consumers of a specific online wine site have high positive perceptions about information quality and service quality,
the impact of the “origin” attribute on perceived risk is greatly
minimized. In particular, information quality has powerful crosslevel interaction effects. This study’s finding supports early website
quality research propositions (Sullivan, 1999; Honeycutt et al.,
1998) that website information quality is the most important factor in reducing perceived risk. When this is applied to the context
of online wine repurchase situations, sites should be designed
to assist consumers with information helpful in describing wine
attributes and particularly those associated with sensory and origin
characteristics. Successful online wine businesses should deliver
customized and value-added services that address consumer needs.
This research finding documents that information and service quality are essential to reduce perceived risk with the online wine
shopping experience.
5.2. Managerial implications
Based upon this study’s findings, managerial implications to
develop effective and successful wine website strategies are now
possible. Gupta et al. (2004) stated that wine is a less amenable
product for electronic retailing because of wine sensory concerns.
Although this study confirmed that consumers having high importance levels of sensory-related attributes might perceive risk with
online wine shopping even if wine websites provide high quality,
it is interesting to note that the power of the “origin” attribute on
perceived risk could be reduced by positive perceptions about wine
website quality. In support of this logic, past studies have emphasized the “origin” attribute (representing region, grape varietal and
type) as an important determinant of wine choice decisions because
118
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
of their strong association with sensory characteristics (Dean, 2002;
Mueller et al., 2010). This is due in part to the fact that types of
grapes (grape varietal) used in making wine influences its taste
(Wang, 2011). Developing surrogate sensory cues using the grape
varietal’s name and origin as examples could potentially assist
consumers with evaluating wine quality prior to purchasing the
product without them necessarily tasting it. This finding could be
used to improve overall website quality which then could reduce
consumer risk for purchasing wine online.
Regarding the three dimensions of website quality (system,
information and service), this study’s findings revealed that system quality does not have a significant effect to mitigate the impact
of the “origin” attribute on perceived risk. However, DeLone and
McLean (2003) emphasized the importance of system quality representing technical capabilities of the website and its usability as a
fundamental factor in the e-service context. Other website quality
literature has asserted that information and service quality are produced on the basis of system quality (DeLone and McLean, 1992;
Xu et al., 2013). Therefore, when considering that system quality
is structurally a precedence factor influencing perceptions about
information and service quality, we should be cautious with interpreting the result regarding an insignificant moderating effect of
system quality. Therefore, the value of system quality should not
be overlooked in the online wine business context.
Results of this study demonstrated that information and service
quality were proven to significantly moderate the impact of the
“origin” attribute upon perceived risk. This finding is similar to
a previous study that revealed website information quality and
service quality are significant to improve consumer purchase intention particularly when applied to the context of food and beverage
businesses (Chiu et al., 2005). Information for consumers lacking
knowledge about wine could help them make purchase decisions. Expanded details could address wine regions, grape varietals
and sensory-related descriptions including appearance, aroma and
taste. Actual opportunities for consumers to interact with wine
makers through virtual presentations that describe the styles and
characteristics of particular wine products would greatly augment
what may be seen as typically uninteresting narrative offering basic
content.
Wine websites should strive to enhance information quality in order to attract and retain consumers. Today’s consumers
expect immediate, accurate and thorough information to assist
with buying wine online. Providing knowledge through education
about wine during the purchasing process could be perceived as a
more desirable step in the direction of website information quality. Websites could also greatly improve information quality by
incorporating technology using participatory features that share
wine-related experiences. Detailed tasting notes, wine reviews,
images and videos are expected to be presented online. Wine websites should involve social media which enables consumers to share
their opinions and comments about wines featured on specific websites.
Beyond delivering well-organized and well-presented information, it is important for wine websites to provide pleasant service
experiences. Consumers now seek website services that are able
to personalize and recommend wine based upon their individual
preferences (Luo and Seyedian, 2003). As well, communication systems designed to provide prompt responses to needs or complaints
of consumers could play a crucial role for improving more desirable levels of wine website service quality. Given the opportunity
to consistently interact with online wine consumers, websites can
customize wine club opportunities to notify club members seeking unique wines. Based upon established wine consumer profiles,
online recommendations could be passed along directly to consumers pertaining to potential online purchases for specialty items.
New vintage releases, opportunities for online purchase of rare
wines, and information about special upcoming events such as
taking a wine-related cruise featuring a winery owner/winemaker
could all add to the perception of websites having outstanding
service quality. Personal follow-up telephone conversations with
consumers tracked on wine websites could help personalize and
customize the service experience. This can establish very positive
and long-term relationships with online wine consumers.
5.3. Limitations and future research
As with all research studies, limitations also exist with this
study. First, although previous HLM research recommendations
regarding sample size adequacy was met in this study, variances
regarding sample size were relatively large when compared to
several other multilevel studies. Regardless, this study obtained
statistically significant results which were consistent with hypothesized expectations that were based upon the existing literature.
This implies that the unequal group sample size issue is less likely
to jeopardize these study results. It is recommended, however, that
future research replicate this study using equal group sample sizes.
Second, this study used an online cross-sectional survey having
purchased wine online during the previous 12 months. Although
this study was in an attempt to minimize sampling issues, findings
are not generalizable to apply to all online wine consumers. A more
representative sampling and a longitudinal design could prove to
generate quite different relationships between wine attributes,
perceived risk, website quality and online repurchase intention.
Therefore, future research should seek more robust results by
using data more representative to online wine consumers. In addition, types of wine websites could be segmented according to
their unique purposes which may possibly offer different systems,
information and services for wine websites. Therefore, consumer
expectations and perceptions about website quality could vary
according to types of wine websites. This presents another opportunity for future research.
Third, this study only focuses on the role website quality plays
on an individual’s perceptions about online wine purchasing. However, there could be other important firm-level variables which
significantly influence individual-level factors. Therefore, future
research should examine other factors that may affect an individual’s attitudes and behaviors regarding online wine purchasing.
Fourth, this study investigated the roles of system quality, information quality and service quality which were placed on the same
link between wine attributes and perceived risk. Thus, it is not certain that the real effects of all three dimensions of website quality
were accurately displayed in this study. Because prior research has
found that consumers perceive website quality according to their
mental sequential process (Xu et al., 2013), it is recommended this
issue be addressed in future studies. Thus, future studies should
attempt to gain a better understanding about the interrelationships among system quality, information quality and service quality
when applied to the web-based wine business context.
References
Ahn, T., Ryu, S., Han, I., 2007. The impact of Web quality and playfulness on user
acceptance of online retailing. Inform. Manag. 44 (3), 263–275.
Ajzen, I., Driver, B.E., 1992. Applied of the theory of planned behavior to leisure
choice. J. Leis. Res. 24 (3), 207–224.
Albert, C., 2004. The impact of switching costs on customer loyalty: a study among
corporate customers of mobile telephone. J. Target. Meas. Anal. Mark. 12 (3),
256.
Alpin, J., 1999. E-commerce in Wine: benefits and barriers. Aust. Wine Ind. J. 14 (6),
87–91.
Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a
review and recommended two-step approach. Psychol. Bull. 103 (3), 411–423.
Andrews, D., Nonnecke, B., Preece, J., 2003. Electronic survey methodology: a case
study in reaching hard-to-involve Internet users. Int. J. Hum.–Comput. Interact.
16 (2), 185–210.
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Bai, B., Law, R., Wen, I., 2008. The impact of website quality on customer satisfaction
and purchase intentions: evidence from Chinese online visitors. IJHM 27 (3),
391–402.
Baker, J., Parasuraman, A., Grewal, D., Voss, G.B., 2002. The influence of multiple store
environment cues on perceived merchandise value and patronage intentions. J.
Mark. 66 (2), 120–141.
Barnes, S.J., Vidgen, R.T., 2002. An integrative approach to the assessment of ecommerce quality. J. Electron. Commer. Res. 3 (3), 114–127.
Bensinger, G., 2012. Wine: The Web’s Final Frontier, Available from: http://online.
wsj.com/news/articles/SB10000872396390444592704578067270510751116
(accessed 12.04.14).
Bernroider, E.W., 2008. IT governance for enterprise resource planning supported
by the DeLone–McLean model of information systems success. Inform. Manag.
45 (5), 257–269.
Bharati, P., Chaudhury, A., 2004. An empirical investigation of decision-making satisfaction in web-based decision support systems. Decis. Support Syst. 37 (2),
187–197.
Bhatnagar, A., Ghose, S., 2004. Segmetation consumers based on the benefits and
risks of Internet shopping. J. Bus. Res. 57 (12), 1352–1360.
Bliese, P.D., 1998. Group size, ICC values and group-level correlations: a simulation.
Org. Res. Methods 1 (4), 355–373.
Boshoff, C., 2007. A psychometric assessment of EQ-QUAL: a scale to measure
eletronic service quality. J. Electron. Commer. Res. 8 (1), 101–114.
Bressolles, G., 2006. Electronic service quality: NetQual-proposition of a measurement scale to commercial websites and moderating effects. Rech. Appl. Mark.
21 (3), 19–45.
Brown, M., Pope, N., Voges, K., 2003. Buying or browsing? An exploration of
shopping orientations and online purchase intention. Eur. J. Mark. 37 (11/12),
1666–1684.
Bruwer, J., Fong, M., Saliba, A., 2013. Perceived risk, risk-reduction strategies (RRS)
and consumption occasions: roles in the wine consumers’ purchase decision.
Asia Pac. J. Mark. Logist. 25 (3), 369–390.
Bruwer, J., Wood, G., 2005. The Australian online wine-buying consumer: motivational and behavioral perspectives. J. Wine Res. 16 (3), 193–211.
Bryk, A.S., Raudenbush, S.W., 1992. Hierarchical linear models: applications and data
analysis. Methods. SAGE, Newbury Park, CA.
Burg, N., 2012. Amazon Tries Selling Wine Online For Third Time, Available from:
http://www.forbes.com/sites/ups/2012/11/16/amazon-will-try-selling-wineonline-for-third-time (accessed 16.01.13).
Cenfetelli, R., Benbasat, I., Al-Natour, S., 2008. Addressing the what and how of online
services: comparing service content and service quality for e-business success.
Inform. Syst. Res. 19 (2), 161–181.
Chang, H.H., Chen, S.W., 2008. The impact of online store environment cues on
purchase intention. Online Inform. Rev. 32 (6), 818–841.
Charters, S., Pettigrew, S., 2005. Is wine consumption an aesthetic experience? J.
Wine Res. 16 (2), 37–52.
Chiou, J.S., Huang, C.Y., Chuang, M.C., 2005. Antecedents of Taiwanese adolescents’
purchase intention toward the merchandise of a celebrity: the moderating effect
of celebrity adoration. J. Soc. Psychol. 145 (3), 317–332.
Chiu, H.C., Hsieh, Y.C., Kao, C.Y., 2005. Website quality: an customer’s behavioral
intention: an exploratory study of the role of information asymmetry. Total Qual.
Manag. 16 (2), 185–197.
Chow, J., 2013. Wine Consultant Charles Curtis on Spotting a Wine Fake, Available from: http://online.wsj.com/news/articles/SB10001424127887324463
604579040671190300050 (accessed 20.01.13).
Couper, M.P., 2000. Web-based surveys: a review of issues and approaches. Public
Opin. Q. 64 (4), 464–494.
Cox, D.F., Rich, S.V., 1964. Perceived risk and consumer decision-making: the case of
telephone shopping. J. Mark. Res. 1, 32–39.
Cronin, J.J., Brady, M.K., Hult, T.M., 2000. Assessing the effects of quality, value, customer satisfaction on customer behavioral intentions in service environment. J.
Retail. 76 (2), 193–216.
Dean, R., 2002. The changing world of the international fine wine market. Aust. N.
Z. Wine Ind. J. May/June, 84–88.
DeLone, W.H., McLean, E.R., 1992. Information systems success: the quest for the
dependent variable. Inform. Syst. Res. 3 (1), 60–95.
DeLone, W.H., McLean, E.R., 2003. The DeLone and McLean model of information
systems success: a ten-year update. J. Manag. Inform. Syst. 19 (4), 9–30.
Dodd, T., Laverie, D.A., Wilcox, J.F., Duhan, D.F., 2005. Differential effects of experience subjective knowledge, and objective knowledge on source of information
used in consumer wine purchasing. J. Hosp. Tour. Res. 29 (3), 3–19.
Featherman, M.S., Pavlou, P.A., 2003. Predicting e-services adoption: a perceived risk
facets perspective. Int. J. Hum.–Comput. Stud. 59 (4), 451–474.
Fornell, C., Larcker, D., 1981. Evaluating structural equation models with unobservable variables and measurement error. J. Mark. Res. 18 (February), 39–50.
Forsythe, S., Liu, C.L., Shannon, D., Gardner, L.C., 2006. Development of a scale to
measure the perceived benefits and risks of online shopping. J. Interact. Mark.
20 (2), 55–75.
Forsythe, S., Shi, B., 2003. Consumer patronage and risk perceptions in Internet
shopping. J. Bus. Res. 56 (11), 867–875.
Gebauer, J., Ginsburg, M., 2003. The US wine industry and the Internet: an analysis of success factors for online business models. Electron. Mark. 13 (1),
59–66.
Glick, W.H., 1985. Conceptualizing and measuring organizational and psychological climate: pitfalls in multilevel research. Acad. Manage. Rev. 10 (3),
601–616.
119
Gupta, A., Su, B., Walter, Z., 2004. An empirical study of consumer switching from
traditional to electronic channels: a purchase-decision process perspective. Int.
J. Electron. Commer. 8 (3), 131–161.
Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis,
5th ed. Printice-Hall, New Jersey.
Hair, J.F., Bush, R.P., Ortinau, D.J., 2000. Marketing Research: A Practical Approach
for the New Millennium. McGraw-Hill, Boston.
Hall, J., Lockshin, L., O’Mahony, G.B., 2001. Exploring the links between wine choice
and dining occasions: factors of influence. Int. J. Wine Mark. 13 (1), 36–53.
Hellier, P.K., Geursen, G.M., Carr, R.A., Rickard, J.A., 2003. Customer repurchase
intention: a general structural equation model. Eur. J. Mark. 37 (11/12),
1762–1800.
Hofmann, D.A., Morgeson, F.P., Gerras, S.J., 2003. Climate as a moderator of the relationship between leader-member exchange and content-specific citizenship:
safety climate as an exemplar. J. Appl. Psychol. 88 (1), 170–178.
Hon, A.H.Y., 2011. Enhancing employee creativity in the Chinese context: the mediating role of employee self-concordance. IJHM 30 (2), 375–384.
Honeycutt, E.D., Flaherty, T.B., Bitner, M.J., 1998. Relational benefits in services industries: the customer’s perspective. J. Acad. Mark. Sci. 26 (2), 101–114.
Hox, J.J., 1998. Multilevel modeling: when and why. In: Balderjahn, I., Mathar, R.,
Schader, M. (Eds.), Classification, Data Analysis, and Data Highways. Springer
Verlag, New York, pp. 147–154.
Huang, W.Y., Schrank, H., Dubinsky, A.J., 2004. Effect of brand name on consumers’
risk perceptions of online shopping. J. Consum. Behav. 4 (1), 40–50.
Jaeger, S.R., Danaher, P.J., Brodie, R.J., 2010. Consumption decisions made in restaurants: the case of wine selection. Food Qual. Prefer. 21 (4), 439–442.
Jagle, S., 2012. Online Wine Welling – Identify Customer Fears. TECHNO, Available
from: http://www.techno-fil.blogspot.com/2012/05/ (accessed 07.01.13).
Johnson, T., Bruwer, J., 2004. Generic consumer risk-reduction strategies (RRS) in
wine-related lifestyle segments of the Australian wine market. Int. J. Wine Mark.
16 (1), 5–34.
Kang, S., Lim, K.J., Kim, M.S., Yang, H.-D., 2012. A multilevel analysis of the effect of
group appropriation on collaborative technologies use and performance. Inform.
Syst. Res. 23 (1), 214–230.
Keown, C., Casey, M., 1995. Purchasing behavior in the Northern Ireland wine market.
Br. Food J. 97 (11), 7–20.
Kettinger, W.J., Park, S.J.S., Smith, J., 2009. Understanding the consequences of information systems service quality on IS service reuse. Inform. Manag. 46 (6),
335–341.
Khalifa, M., Liu, V., 2007. Online consumer retention: contingent effects of online
shopping habit and online shopping experience. Eur. J. Inform. Syst. 16 (6),
780–792.
Kim, D.J., Ferrin, D.L., Rao, H.R., 2008. A trust-based consumer decision-making
model in electronic commerce: the role of trust, perceived risk, and their
antecedents. Decis. Support Syst. 44 (2), 544–564.
Kim, M., Kim, J.H., Lennon, S.J., 2006. Online service attributes available on apparel
retail websites: an E-S-QUAL approach. Manag. Serv. Qual. 16 (1), 51–77.
Kim, S., Stoel, L., 2004. Dimensional hierarchy of retail website quality. Inform.
Manag. 41 (5), 629–633.
Klein, K.J., Kozlowski, S.W., 2000. From micro to meso: critical steps in conceptualizing and conducting multilevel research. Org. Res. Methods 3, 211–236.
Lacey, S., Bruwer, J., Li, E., 2009. The role of perceived risk in wine purchase decisions
in restaurants. Int. J. Wine Bus. Res. 21 (2), 99–117.
Lee, Y., Kozar, K.A., 2006. Investigating the effect of website quality on e-business
success: an analytic hierarchy process (AHP) approach. Decis. Support Syst. 42
(3), 1383–1401.
Lee, K.S., Tan, S.J., 2003. E-retailing versus physical retailing: a theoretical model and
empirical test of consumer choice. J. Bus. Res. 56 (11), 877–885.
Li, Y.N., Tan, K.C., Xie, M., 2002. Measuring web-based service quality. Total Qual.
Manag. 13 (5), 685–700.
Liang, A.R.D., Lim, W.M., 2011. Exploring the online buying behavior of specialty
food shoppers. IJHM 30 (4), 855–865.
Lim, N., 2003. Consumers’ perceived risk: sources versus consequences. Electron.
Commer. Res. Appl. 2 (3), 216–228.
Lin, H.F., 2007. The impact of website quality dimensions on customer satisfaction
in the B2C e-commerce context. Total Qual. Manag. 18 (4), 363–378.
Lin, H.F., 2008. Determinants of successful virtual communities: contributions from
system characteristics and social factors. Inform. Manag. 45 (8), 522–527.
Liu, C.K., Arnett, P., 2000. Exploring the factors associated with web site success in
the context of electronic commerce. Inform. Manag. 38 (1), 23–33.
Long, M., McMellon, C., 2004. Exploring the determinants of retail service quality on
the Internet. J. Serv. Market. 18 (1), 78–90.
Loureiro, M.L., 2003. Rethinking new wines: implications of local and environmentally friendly labels. Food Policy 28 (5–6), 547–560.
Luke, D.A., 2004. Multilevel Modeling Quantitative Applications in the Social Science.
Sage Publishing, Newbury Park, CA.
Luo, X., Seyedian, M., 2003. Contextual marketing and customer-orientation strategy for e-commerce: an empirical analysis. Int. J. Electron. Commer. 8 (2),
95–118.
Lyons, J.S., Terry, P., Martinovich, Z., Peterson, J., Bouska, B., 2001. Outcome trajectories for adolescents in residential treatment: a statewide evaluation. J. Child
Family Stud. 10 (3), 333–345.
Menon, S., Kahn, B., 2002. Cross-category effects of induced arousal and pleasure on
the Internet shopping experience. J. Retail. 78 (1), 31–40.
Miyazaki, A.D., Fernandez, A., 2001. Consumer perceptions of privacy and security
risks for online shopping. J. Consum. Aff. 35 (1), 27–44.
120
M. Cho et al. / International Journal of Hospitality Management 43 (2014) 108–120
Mueller, S., Osidacz, P., Francis, L., Lockshin, L., 2010. Combining discrete choice and
informed sensory testing in a two-stage process: can it predict wine market
share? Food Qual. Prefer. 21 (7), 741–754.
Mueller, S., Szolnoki, G., 2010. The relative influence of packaging, labeling, branding
and sensory attributes on liking and purchase intent: consumers differ in their
responsiveness. Food Qual. Prefer. 21 (7), 774–783.
Ness, L., 2006. Creating effective winery websites: is yours worthy? Wines Vines
June, 84–89.
Nowak, L.I., Newton, S., 2008. Using winery web sites to launch relationships with
Millennials. Int. J. Wine Bus. Res. 20 (1), 53–67.
Nunnally, J.C., 1978. Psychometric Theory, 2nd ed. McGraw-Hill, New York.
Oczkowski, E., 2001. Hedonic wine price functions and measurement error. Econ.
Rec. 77 (239), 374–382.
Oppenheimer, D.M., Meyvis, T., Davidenko, N., 2009. Instructional manipulation
checks: detecting satisficing to increase statistical power. J. Exp. Soc. Psychol.
45 (4), 867–872.
Palmer, J.W., 2002. Website usability, design, and performance metrics. Inform. Syst.
Res. 13 (2), 151–167.
Parasuraman, A., Zeithaml, V.A., Malhotra, A., 2005. E-S-QUAL, a multiple-item scale
for assessing electronic service quality. J. Serv. Res. 7 (3), 213–233.
Park, C., Jun, J.K., 2003. A cross-cultural comparison of Internet buying behavior:
effects of Internet usage, perceived risks and innovativeness. Int. Mark. Rev. 20
(5), 534–553.
Park, C.H., Kim, Y.G., 2003. Identifying key factors affecting consumer purchase
behavior in an online shopping context. IJRDM 31 (1), 16–29.
Pavlou, P.A., 2003. Consumer acceptance of electronic commerce: integrating trust
and risk with the technology acceptance model. Int. J. Electron. Commer. 7 (3),
101–134.
Pfanner, E., 2013. Buying wine on the web. The New York Times, Available
from: http://www.nytimes.com/2013/02/02/dining/02iht-wine02.html? r=0
(accessed 14.02.13).
Phau, I., Poon, S.M., 2000. Factors influencing the types of products and services
purchased over the Internet, Internet research. Electron. Netw. Appl. Policy 10
(2), 102–113.
Pires, G.D., Stanton, J., Eckford, A., 2004. Influences on the perceived risk of purchasing online. J. Consum. Behav. 4 (2), 118–131.
Quester, P.G., Smart, J., 1996. Product involvement in consumer wine purchases: its
demographic determinants and influence on choice attributes. Int. J. Wine Mark.
8 (3), 37–56.
Quester, P.G., Smart, J., 1998. The influence of consumption situation and product
involvement over consumers’ use of product attribute. J. Consum. Mark. 15 (3),
220–238.
Raudenbush, S.W., 1998. Educational applications of hierarchical linear models: a
review. J. Educ. Stat. 13 (2), 85–116.
Santos, J., 2003. E-service quality: a model of virtual service quality dimensions.
Manag. Serv. Qual. 13 (3), 233–246.
Santos, J.F., Ribeiro, J.C., 2012. The Portuguese online wine buying consumer: characteristics, motivations and behavior. EuroMed J. Bus. 7 (3),
294–311.
Schiffman, L.G., Kanuk, L.L., 2006. Consumer Behavior, 8th ed. Prentice Hall International, Englewood Cliffs, NJ.
Siegrist, M., Cousin, M.E., 2009. Expectations influence sensory experience in a wine
tasting. Appetite 52 (3), 762–765.
Stefalo, T., 2011. Why Buy Wine Online, Available from: http://www.best4wine.
co.uk/tag/buy-wine-online/page/2/ (accessed 16.09.13).
Stricker, S., Mueller, R.A.E., Sumner, D.A., 2001. Wine on the web. Rapid appraisal of
websites by wineries and wine merchants. In: Dep. of Agricultural and Resource
Economics, University of California, Davis, CA.
Sullivan, J., 1999. What are the functions of corporate home pages? J. World Bus. 34
(2), 193–210.
Swartzberg, M., Solomon, J.F., Berkman, M., 2000. Clicking on Wine, Salomon Smith
Barney Equity Research, Beverages Division (March 7).
Tan, S.J., 1999. Strategies for reducing consumers’ risk aversion in Internet shopping.
J. Consum. Mark. 16 (2), 163–178.
Taylor, J.W., 1974. The role of risk in consumer behavior. J. Mark. 38 (2), 54–60.
Tsai, H.T., Huang, H.C., 2007. Determinants of e-repurchase intentions: an integrative
model of quadruple retention drivers. Inform. Manag. 44 (3), 231–239.
Tsai, W.H., Shaw, M.J., Fan, Y.W., Liu, J.Y., Lee, K.C., Chen, H.C., 2011. An empirical investigation of the impacts of internal/external facilitators on the project
success of ERP: a structural equation model. Decis. Support Syst. 50 (2), 480–490.
Tzimitra-Kalogianni, I., Papadaki-Klavdianou, A., Alexaki, A., Tsakiridou, E., 1999.
Wine routes in Northern Greece: consumer perceptions. Br. Food J. 101 (11),
884–892.
Udo, G.J., Bagchi, K.K., Kirs, P.J., 2010. An assessment of customers’ e-service quality
perception, satisfaction and intention. Int. J. Inform. Manag. 30 (6), 481–492.
Verhoef, P.C., Langerak, F., 2001. Possible determinants of consumers’ adoption of
electronic grocery shopping in the Netherlands. J. Retail. Consum. Serv. 8 (5),
275–285.
Vitell, S.J., Kinghapakdi, A., Thomas, J., 2001. Consumer ethics: an application and
empirical testing of the Hunt–Vitell theory of ethics. J. Consum. Mark. 18 (2),
153–178.
Vizu, 2007. Online Wine Consumer Survey-Market Research Report. Vizu Corporation, Available from: http://ww1.prweb.com/prfiles/2007/09/17/246499/
0 OnlineWineMarketResearch.pdf (accessed 14.04.14).
Wang, H.Y., 2011. Exploring the factors of gastronomy blogs influencing readers’
intention to taste. IJHM 30 (3), 503–514.
Wang, Y.S., 2008. Assessing e-commerce systems success: a re-specification and
validation of the DeLone and McLean model of IS success. Inform. Syst. J. 18 (5),
529–557.
Whitener, E.M., 2001. Do high commitment human resource practices affect
employee commitment? A cross-level analysis using hierarchical linear modeling. J. Manag. 27 (5), 515–535.
Wine Institute, 2013. Department of Commerce, Alcohol and Tobacco Tax and
Trade Bureau, Available from: http://www.ttb.gov/statistics/13winestats.shtml
(accessed 23.09.13).
Wine Market Council, 2012. Wine Market Council Consumer Tracking Study on
the U.S. Wine Market, Available from: http://www.winemarketcouncil.com
(accessed 14.04.14).
Wiseman, A.E., Ellig, J., 2007. The politics of wine: trade barriers, interest groups,
and the commerce clause. J. Politics 69 (3), 859–875.
Wolfinbarger, M., Gilly, M.C., 2001. Shopping online for freedom, control, and fun.
Calif. Manag. Rev. 43 (2), 34–56.
Wolfinbarger, M., Gilly, M.C., 2003. eTailQ: dimensionalizing, measuring and predicting etail quality. J. Retail. 79 (3), 183–198.
Woltman, H., Feldstain, A., MacKay, J.C., Rocchi, M., 2012. An introduction to hierarchical linear modeling. Tutor. Quant. Methods Psychol. 8 (1), 52–69.
Wu, F., Mahajan, S., Balasubramanian, S., 2003. An analysis of e-business adoption
and its impact on business performance. J. Acad. Mark. Sci. 31 (4), 425–447.
Xu, J., Benbasat, I., Cenfetelli, R.T., 2013. Integrating service quality with system and
information quality: an empirical test in the e-service context. MIS Q. 37 (3),
777–794.
Yoo, B., Donthu, N., 2001. Developing and validating a multidimensional consumerbased brand equity scale. J. Bus. Res. 52 (1), 1–14.
Zeithaml, V.A., Parasuraman, A., Malhotra, A., 2002. Service quality delivery through
websites: a critical review of extant knowledge. J. Acad. Mark. Sci. 30 (4),
362–375.