The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1468-4527.htm

OIR
32,6 The impact of online store
environment cues on purchase
intention
818
Trust and perceived risk as a mediator
Refereed article received
17 January 2008
Hsin Hsin Chang and Su Wen Chen
Approved for publication Department of Business Administration, National Cheng Kung University,
17 May 2008 Taiwan

Abstract
Purpose The purpose of this paper is to investige whether online environment cues (web site
quality and web site brand) affect customer purchase intention towards an online retailer and whether
this impact is mediated by customer trust and perceived risk. The study also aimed to assess the
degree of reciprocity between consumers trust and perceived risk in the context of an online shopping
environment.
Design/methodology/approach The study proposed a research framework for testing the
relationships among the constructs based on the stimulus-organism-response framework. In addition,
this study developed a non-recursive model. After the validation of measurement scales, empirical
analyses were performed using structural equation modelling.
Findings The findings confirm that web site quality and web site brand affect consumers trust
and perceived risk, and in turn, consumer purchase intention. Notably, this study finds that the web
site brand is a more important cue than web site quality in influencing customers purchase intention.
Furthermore, the study reveals that the relationship between trust and perceived risk is reciprocal.
Research limitations/implications This study adopted four dimensions technical adequacy,
content quality, specific content and appearance to measure web site quality. However, there are still
many competing concepts regarding the measurement of web site quality. Further studies using other
dimensional measures may be needed to verify the research model.
Practical implications Online retailers should focus their marketing strategies more on
establishing the brand of the web site rather than improving the functionality of the web site.
Originality/value This study proposed a non-recursive model for empirically analysing the link
between web site quality, web site brand, trust, perceived risk and purchase intention towards the
online retailer.
Keywords Worldwide web, Internet, Trust, Buying behaviour, Risk analysis
Paper type Research paper

Online Information Review
Vol. 32 No. 6, 2008
pp. 818-841
q Emerald Group Publishing Limited
1468-4527
DOI 10.1108/14684520810923953
Introduction
With annual web-generated sales in the billions of dollars (Arcand et al., 2007), web
sites are valuable retail outlets because of their wide domestic and international reach,
low cost, constant innovation, and order taking and customer feedback facilities
(Eroglu et al., 2001). However, unlike offline retail stores, the inability to interact with a
salesperson and the merchandise, as well as the reliance on electronic payment
methods increase perceived risk with regard to online shopping (Casalo et al., 2007).
Consumers thus look for cues from the online environment to make sure that they will
not experience a loss when making a purchase online.
 A web site is perhaps the only way an online store communicates with its customers Online store
(Chen and Dhillon, 2003). Thus web site quality serves as the store atmosphere (Ethier environment
et al., 2006) and accordingly is a trustworthiness cue (Corritore et al., 2003), especially at
the beginning of transactions (McKnight et al., 2004). However, providing good web
cues
site quality may be not enough to motivate customers to adopt online shopping (Cho,
2004), since consumers dislike dealing with unknown vendors (Lim, 2003). For
example, several researchers argue that when dealing with unknown vendors, 819
consumers are more concerned about privacy and are less likely to trust them (e.g.
Gefen, 2000; Walczuch and Lundgren, 2004). Thus web site brand can present
intangible cues deliberately used to build consumer trust in an online retailers
competence, integrity and benevolence since a branded web site can reassure
customers that the business is reliable, safe and trustworthy (Hashim and Murphy,
2007). For example, as the first online bookstore, Amazon has received substantial
recognition in online book retailing and has been persistently the leader in the online
book retail industry.
Recently, the effect of online environment cues on consumer purchase behaviour has
received increasing managerial and research attention. (Chang and Wang, 2008; Davis
et al., 2007; Eroglu et al., 2001, 2003; Ethier et al., 2006). However, although the brand is
an important cue to influence consumers online purchase intention (Rowley, 2004;
Yoon, 2002), few studies have examined how it affects consumers online shopping
behaviour.
Furthermore, although most studies in environmental psychology have focused on
the pleasure-arousal-dominance (PAD) dimensions of emotional response as expected
reactions to environmental stimuli, this approach has been criticised as being too
narrow in scope and not capturing much of the effect of site atmosphere (Eroglu et al.,
2001, 2003). Instead, the concept of trust and perceived risk can be viewed as
psychological states (Cho and Lee, 2006; Rousseau et al., 1998) and have been identified
as critical mediators that influence customers online purchase intention (Chau et al.,
2007; Chen and Barnes, 2007; Lim, 2003; Lin, 2007; Mitchell, 1999; Wang and Emurian,
2005). For example, Chang et al. (2005) developed an exhaustive literature-derived
model of online shopping. They then derived the relationships among antecedents of
the determinants of online shopping and demonstrated that trust and perceived risk
are the most researched mediator variables. Therefore, this study adopts both trust and
perceived risk as mediator variables in the research framework. Moreover, there has
been much confusion about the relationships between trust and perceived risk (Lee and
Lee, 2007; Lim, 2003). Mayer et al. (1995) insisted that it is unclear whether trust comes
before perceived risk or vice versa. Mitchell (1999) reviewed the literature on consumer
perceived risk and considered the relation between trust and perceived risk to be
non-recursive. However, the reciprocal relation of trust and perceived risk has not been
studied empirically.
Addressing these research gaps, this study developed a non-recursive research
model about the impact of web site quality and brand on consumers trust and
perceived risk, and in turn, purchase intention. Notably, this study considered the
relationships between trust and perceived risk to be non-recursive. The purpose of the
research was:
OIR .
To investigate whether perceptions of web site quality and brand significantly
32,6 impact consumers purchase intention towards an online retailer and whether
this impact is mediated by consumers trust and perceived risk.
.
To clarify which online environmental cue (web site quality or web site brand) is
more influential in affecting consumers purchase intention towards an online
retailer.
820 .
To assess the degree of reciprocity between consumers trust and perceived risk
in the context of an online environment.

Conceptual framework and literature review

Conceptual framework
Figure 1 presents a research framework based on the stimulus-organism-response
(S-O-R) paradigm (Eroglu et al., 2001, 2003). Within the S-O-R framework, stimulus is
conceptualised as an influence that affects internal, organismic states of the individual.
The organism is represented by cognitive and affective intermediary states and
processes that mediate the relationships between the stimulus and the individuals
responses. The response represents the final outcome, the approach or avoidance
behaviours of the consumer. Baker (1986) presented a typology categorising the
elements of store environment into three categories: social, design and ambient factors.
Social factors relate to salespeople in the store, ambient factors relate to the non-visual
elements of a stores environment and design factors are visual in nature (e.g. layout,
colour, cleanliness, clusters, space, etc.). Bitners (1992) typology included ambient cues
(those cues that affect the five senses), layout and functionality (store arrangement and
ability to facilitate consumer goals), and signs, symbols and artifacts (signals that
communicate information to the shopper). Sherman et al. (1997) suggested that store
image is another element of store environment.

Figure 1.
Conceptual framework
In the context of online shopping, the stimulus is the web site atmosphere as it affects Online store
the internal states of the consumer. Researchers have pointed out that online store environment
atmospherics include all the cues used to design the web site and its layout, with
examples including the background colour and pattern, hyperlinks, icons, overall cues
colour scheme, typeface and web borders (Chang and Wang, 2008; Davis et al., 2007;
Eroglu et al., 2001). Based on previous studies, this study defined the stimulus as the
cues that are tangible (e.g. a web sites technical adequacy, content quality, specific 821
content and appearance) and intangible (e.g. web site awareness and web site image) to
internet users.
Additionally, according to various researchers views (e.g. Bhattacherjee, 2002; Cho
and Lee, 2006), trust and perceived risk are psychological states that are cognitive or
affective episodes that fluctuate with situational contexts (Limayem and Khalifa, 2000)
and may be influenced by an individuals interaction with a situation. Hence, this study
treated trust and perceived risk as organism variables that are affected by stimuli.
Accordingly, this study posited that web site quality and web site brand (stimuli)
positively affect consumers trust and negatively affect perceived risk (organism),
which in turn may affect consumers purchase intention towards the online retailer
(response). Furthermore, based on previous studies (Chang, 2008; Lim, 2003; Mitchell,
1999), the relationships between trust and perceived risk were hypothesised to be
non-recursive (bidirectional).

Web site quality

Web site quality is defined here as users evaluations of whether a web sites features
meet users needs and reflect the overall excellence of the web site (Aladwani and
Palvia, 2002). Internet users are classified into internet shoppers (those who have made
purchases on the internet) and internet browsers (those who have browsed online for
products or services but have not made purchases) (Forsythe and Shi, 2003). The
quality of the web site affects the users view of the site because it is the portal through,
which transactions are conducted (Ahn et al., 2007).
Several dimensions of web site quality have been identified in the literature. Most
studies have evaluated web site quality from the shoppers perspective and developed
multi-item scales to measure the entire online buying experience that includes the
pre-purchase and post-purchase experiences of the customer (Negash et al., 2003;
Wolfinbarger and Gilly, 2003). For example, Bauer et al. (2006) developed a transaction
process-based scale for measuring e-service quality (eTransQual) and identified five
eTransQual dimensions: functionality/design, enjoyment, process, reliability and
responsiveness. Aladwani and Palvia (2002) set out to develop a sound instrument to
measure web site quality from the users perspective and identified four underlying
dimensions: technical adequacy, content quality, specific content and appearance.
This study primarily focused on the web interface and is related to the internet
users browsing experience and not pre- and post-purchase issues. The instrument
developed by Aladwani and Palvia (2002) was considered appropriate for this study in
view of its sound psychometric properties and its focus on the web interface rather
than on the entire online buying experience. Hence, this study adopted the four
dimensions of web site quality proposed by Aladwani and Palvia (2002) mentioned
above. Technical adequacy means that appropriate technologies have been adopted by
the web retailer, such as ease of navigation, speed of page loading, valid links and so
OIR on. Content quality deals with attributes such as information usefulness, completeness,
32,6 accuracy and so on. Specific content refers to finding specific details about
product/services, customer support, privacy polices and other important information.
Appearance refers to the visual attractiveness of a web site, such as colours, fonts and
multimedia features (Aladwani, 2006; Aladwani and Palvia, 2002; Chang and Wang,
2008; Ranganathan and Ganapathy, 2002).
822
Web site brand
A brand is a name and/or symbol-like logo, trademark and package design that
uniquely identifies the products or services of a seller, and differentiates them from
those of its competitors (Aaker, 1991). In the service industry, a companys name is the
primary brand (Berry, 2000; Davis et al., 2007). In the context of service branding,
researchers have noted the importance of brand name in consumer evaluations of
brands, as it assists the consumer in recalling brand benefits (Grace and OCass, 2005;
Keller, 2003). In fact, it is argued that the brand name is more important when there is a
lack of available information (as is often the case with services) due to the brand name
becoming a surrogate for missing attribute information (Grace and OCass, 2005).
Although there is limited research in the area of web site brands, it is reasonable to
argue that the name of the retailers web site is its brand.
Aaker (1996) defined brand equity as a set of assets linked to a brands name and
symbol that adds to the value provided by a product or service to a firm and/or that
firms customers. Keller (1993) presented a knowledge-based framework for creating
brand equity, in which knowledge is composed of two major dimensions: brand
awareness and brand image. Awareness is composed of brand recall and recognition,
while image is composed of various associations with the brand.
Just like a traditional concept of brand awareness, web site awareness is defined
here as the ability of a potential buyer to recognise or recall a specific web site. Web site
recognition reflects familiarity gained from past exposure. Web site recall refers to the
ability to remember a web site. Familiarity is not the same as experience over time.
Experience results from interaction with a process, while familiarity is the result of
mere exposure to a person, a store or an event (Walczuch and Lundgren, 2004).
Similarly, web site image is defined as perceptions about a web site name as
reflected by the web site associations held in consumer memory. According to Biel
(1992), brand image has three contributing sub-images that of the maker (corporate
image), of the product/service, and of the users. Brand image could come from a variety
of sources, including consumer experience, marketing communications and/or word of
mouth (Romaniuk and Sharp, 2003).

Trust
Trust is a psychological state (Rousseau et al., 1998) and a multi-faceted concept that
incorporates cognitive and affective trust (Johnson and Grayson, 2005; Lewis and
Weigert, 1985; Riegelsberger et al., 2003). A trustor can have cognitive trust, which is
good rational reasons why the object of trust merits trust, or affective trust that is
motivated by strong positive feelings towards that which is trusted (Lewis and
Weigert, 1985, p. 972; Corritore et al., 2003). Lewis and Weigert (1985) argued that
cognitive trust is more typical at the macro level in large settings or societies, whereas
affective trust is more typical in primary, close-knit groups or situations. Cognitive
trust is also defined as customers confidence or willingness to rely on a trustees Online store
competence and reliability (Johnson and Grayson, 2005), while affective trust is based environment
on the trustees benevolence (Riegelsberger et al., 2003). Typically, cognition and
affection are intertwined. For example, consumers trust in gardnersnet.com is cues
facilitated by the reliable information about plants and planting techniques as well as
the rich green background scheme on the web site and the colorful pictures of new
varieties of flowers and vegetables (Corritore et al., 2003). The consumers trust thus 823
has both cognitive and affective elements. Accordingly, trust is considered a cognitive
and affective state in the conceptual framework presented here.

Perceived risk
The theory of perceived risk has been applied to explain consumers behaviour in
decision making since the 1960s. Mitchell (1999) suggested that perceived risk is
powerful at explaining consumers behaviour because consumers are more often
motivated to avoid mistakes than to maximise utility in purchasing. The definition of
perceived risk has changed since online transactions have become popular. In the past,
perceived risks were primarily regarded as fraud and product quality (Wu and Wang,
2005). Today, perceived risk refers to certain types of financial, product performance,
social, psychological, physical and time risks when consumers make transactions
online (Boksberger et al., 2007; Chang, 2008; Corbitt et al., 2003; Lim, 2003; Mitchell,
2001; Smith and Sivakumar, 2004). According to Kim et al. (2007), perceived risk is
defined as a consumers belief about the potential uncertain negative outcomes from
the online transaction. As perceived risk is an individuals biased assessment of a risk
situation, its assessment is highly dependent on the individuals psychological and
situational characteristics (Cho and Lee, 2006). Accordingly, this study adopts
perceived risk as an organism variable affected by online environment cues.

Hypotheses development
Web site quality, trust and perceived risk
Web site quality serves as the store atmosphere (Ethier et al., 2006). A well-designed
web site can enhance the probability of a favourable impression as the viewer responds
to visual cues, and a viewer with a favourable impression of a web site is more likely to
become a customer (Albert, 2004; Napier et al., 2001). McKnight et al. (2004) indicated
that perceived web site quality relates positively to both trusting belief and trusting
intentions because using the web site provides the first experiential taste of the
vendors presence, solidifying initial impressions. Therefore, if consumers perceive
that the web site is high quality, they are likely to have high trusting belief about the
online retailers competence, integrity and benevolence, and will develop a willingness
to purchase (McKnight et al., 2004; McKnight et al., 2002).
Additionally, Lim (2003) argued that the consumers perceived risk is
technology-related and includes issues such as download delays, limitations in the
interface, search problems, inadequate measurement of web application success,
security weakness and a lack of internet standards. Consumers appreciate simplicity or
clear design in e-commerce web sites because they reduce the perceived risks of wasted
time, deception and frustration, and because customers may get annoyed when they
see the interface design and format of interface elements varying among the different
OIR pages of a web site (Wang and Emurian, 2005). Hence, this study hypothesised the
32,6 following:
H1. Perceptions of a retailers web site quality positively affect trust towards the
online retailer.
H2. Perceptions of a retailers web site quality negatively affect perceived risk
824 towards the online retailer.

Brand of retailers web site, trust and perceived risk

A web site commanding high consumer awareness will influence trust and satisfaction
(Yoon, 2002), since most consumers assume that a well-known web site has better
capabilities to fulfil their demands (Harridge-March, 2006; Jari and Heikki, 2007;
Koufaris and Hampton-Sosa, 2004). According to a study by Cheskin Research, brand
is at the top of the list of six marketplace fundamentals for building and maintaining
trust on the internet (the others are navigability, fulfilment, presentation, technology
and seals of approval) (Ibeh et al., 2005). Therefore, web site brand recognition is a
powerful factor that leads to consumer trust (Chen and Dhillon, 2003). Furthermore,
Bhattacherjee (2002) postulated that familiarity is a predictor for trust in online firms
and for the consumers willingness to undertake a transaction. Fam et al. (2004)
indicated that a well-known web site brand image will have an effect on consumers
overall purchase decisions by influencing their trust towards a specific web site.
Several researchers have pointed out that trust serves a pivotal role in attracting
customers to shop (Hoffman et al., 1999; Reichheld and Schefter, 2000).
According to the consumer decision-making model proposed by Blackwell et al.
(2001), internal information retrieved from memory (e.g. familiarity, prior experience)
could reduce perceived risk. For example, customers perceive a low level of risk in
Amazon.com because it is a well-known internet vendor and was among the pioneers
of e-commerce (Gefen, 2000; Lim, 2003). In a consumer perceived risk study, Lim (2003)
found that consumers dislike dealing with unknown vendors because they are afraid
that their credit cards will be misused, while many consumers perceive less risk in a
reputable business and rely on references from other people instead of individual trial
and error. Therefore, a well-established web site brand could reduce customers
perceived risks regarding finances and product performance in buying products that
are difficult to evaluate prior to purchase. Although there is limited research in the
area, it is reasonable to expect that web site brand could positively affect online trust
and negatively affect perceived risk, and in turn increase purchase intention towards
the online retailer. Hence, this study hypothesised the following:
H3. Perceptions of a retailers web site awareness and image positively affect trust
towards the online retailer.
H4. Perceptions of a retailers web site awareness and image negatively affect
perceived risk towards the online retailer.
H6. Consumers trust positively affects purchase intention towards the online
retailer.
H7. Perceived risk negatively affects purchase intention towards the online
retailer.
The relations between trust and perceived risk Online store
The literature shows that researchers have different views about the relations between environment
trust and perceived risk. There are the four types of relations identified by researchers
(Gefen, 2000; Lim, 2003). In the first type, perceived risk moderates the relation between cues
consumers trust and purchase intention. For example, Stewart (1999) considered
perceived risk to be a moderating factor on the relation between consumers trust and
their willingness to purchase products from internet vendors. 825
In the second type of relation, perceived risk precedes consumers trust. For
example, de Ruyter et al. (2001) found that a low perceived risk level results in higher
customer trust in, perceived quality of and intention to use an e-service than a high-risk
level. Corbitt et al. (2003) proposed an online B2C (business-to-consumer) perceived
trust model and argued that perceived risk negatively affects perceived trust.
Similarly, Corritore et al. (2003) contended that perception of risk affects consumer
trust in the context of online shopping. Lee and Lee (2007) found that perceived risk
negatively affected trust in a mobile banking empirical study.
In the third relation, trust precedes perceived risk. For example, Pavlou (2003)
posited that trust in the online retailer affects consumers perceived risk of transaction
and their intention to undertake a transaction. Grabner-Krauter and Kaluscha (2003)
argued that consumers seek to reduce the uncertainty and complexity of transactions
and relationships in electronic markets by applying trust. In a recent study, Teo and
Liu (2007) found that in e-commerce, consumers trust has a negative relationship with
perceived risk
The fourth type of relation between the two factors is non-recursive. For example,
Mitchell (1999) reviewed the literature on consumer-perceived risk over the past 30
years and concluded that perceived risk is a necessary antecedent for trust to be
operative and that an outcome of trust building is a reduction in the perceived risk of
the transaction. Although the reciprocal relation of the consumers trust and perceived
risk has not yet been empirically studied, based on previous studies, it is reasonable to
expect that the relationships between trust and perceived risk are reciprocal. Thus, this
study hypothesised the following:
H5. The relationships between trust and perceived risk are non-recursive.
H5a. Consumers trust towards a specific online retailer negatively affects
perceived risk towards the online retailer.
H5b. Perceived risk towards a specific online retailer negatively affects consumers
trust towards the online retailer.

Control variable
Trust propensity (dispositional trust) is the general willingness to trust other people
and a measure of an individuals tendency to trust or distrust (Gefen, 2000; Gefen and
Straub, 2004; Jari and Heikki, 2007). It is influenced by cultural background,
personality type and previous experiences. Thus internet retailers cannot influence
dispositional trust by applying specific trust building strategies (Corritore et al., 2003).
Researchers have suggested that dispositional trust should be included in empirical
studies as a control variable of the trust model (Gefen and Straub, 2004).
Similarly, risk propensity is defined as an individuals general tendency towards
either taking or avoiding risk within a particular kind of decision context (Cho and Lee,
OIR 2006; Sitkin and Pablo, 1992). Risk propensity is a general personality trait that causes
32,6 individuals to demonstrate consistent risk-seeking or risk-averse tendencies across a
variety of situations. For example, a risk-averse decision-maker is more likely to
recognise and weigh negative outcomes, thereby overestimating the probability of a
loss relative to the probability of a gain. This overestimation will result in higher risk
perceptions (Mullins and Forlani, 2005). To validate the proposed model, this study
826 controlled for the effect of consumer personality. Accordingly, trust propensity and
risk propensity were added as control variables in this study.

Method
Questionnaire development
The survey questionnaire was composed of three sections. In Section 1, each
respondent was asked to circle an online retailer web site that they frequently browsed.
Unlike some B2C studies where respondents were generally asked to rate particular
popular web sites, respondents in this study were asked to evaluate a web site they had
already frequently browsed. This approach ensured that the respondent had both
knowledge and experience of the web site they were evaluating.
In Section 2, the measures were designed to evaluate the feelings respondents held
towards a specific web site. The questionnaire used seven-point Likert-type items,
anchored by 1 strongly disagree and 7 strongly agree, adopted from existing
scales. Perceived web site quality was measured by 25 item measures adapted from
Aladwani and Palvia (2002). The scale was further examined by Liao et al. (2006), who
found it to have high reliability. Web site awareness and web site image were assessed
by four item measures each adopted from Aaker (1996) and Davis et al. (2007)). Trust
was assessed by nine item measures modified from Gefen and Straub (2004). The scale
was further examined by Hwang and Kim (2007), who found them to have high
reliability. Perceived risk was assessed by six item measures modified from Corbitt
et al. (2003). Purchase intention was assessed by three item measures modified from
Pavlou (2003). Furthermore, trust propensity was assessed by three item measures
(Chen and Barnes, 2007; Koufaris and Hampton-Sosa, 2004). Risk propensity was
assessed by three item measures (Cho and Lee, 2006).
Section 3 of the questionnaire included demographic data, such as gender, age, level
of education and occupation, shopping experience and internet experience.

Pilot test
The preliminary instrument was pilot-tested with a convenient sample. There were 156
responses, of which 122 were complete, giving a response rate of 78 per cent. The
results of the pilot test were evaluated using Cronbachs reliability and factor analysis.
Cronbachs alpha indicator was used to assess the initial reliability of the scales. The
standard lower bound for Cronbachs alphas is 0.70 (Hair et al., 1998). The item-to-total
correlation was used to improve the level of the Cronbachs alpha, considering a
minimum value of 0.5 (Churchill, 1995). Items that did not significantly contribute to
the reliability were eliminated. A factor analysis was then performed to examine
whether the items produced the expected number of factors and whether the individual
items were loaded on their appropriate factor as expected. Hair et al. (1998) considered
a measure to be significant if its factor loading is greater than 0.50 when the sample
size is 120. This criterion was adopted to examine the item loadings of all measures.
The measurement was then refined by removing items that did not load significantly Online store
onto the underlying constructs. In addition, modifications to refine the instrument were environment
made based on suggestions by respondents. The final 48 measures obtained from the
pilot test were used as indicator variables for the main study. A final text version of the cues
questionnaire items is presented in the Appendix.

Sampling plan and data collection 827

This study conducted a web-based survey to maintain respondents anonymity and
overcome time and place constraints, thus helping our study to reach respondents more
easily than with other survey methods such as interviews (personal and telephone) and
other self-administered survey techniques (Wang and Emurian, 2005; Wilde et al.,
2004).
The internet questionnaire was hosted by Chungwa Telecom Co. Ltd. The
questionnaire collection ran continuously for the six-week duration of the survey
period. To solicit a pool of respondents who would be as close to the general public of
internet users as possible, the link to the survey was distributed through university
listservs, online discussion boards and personal email contacts. Furthermore, the
message was also subsequently repeatedly posted on various online discussions
boards to encourage more responses. The participants were volunteers who were also
interested in the research topic. Since the survey did not target specific individuals,
there is no response-rate calculation.
The main survey had 672 responses. We eliminated 44 respondents who were
obviously unconcerned (i.e. giving the same rating for all items), and 628
questionnaires were retained for analysis. Approximately 54 per cent of the
respondents were male. Of the respondents, 84 per cent were educated to
institute/college level and above, 24 per cent were students and 71 per cent were
workers. With regard to age, 10 per cent were aged below 20, 62 per cent were aged
between 20 and 34, and 21 per cent were aged between 35 and 44. According to
TWNICs (2008) report regarding the age distribution of the population of internet
users, 20 per cent of internet users are aged between 12 and 20, 40 per cent are aged
between 21 and 35, and 23 per cent are aged between 36 and 45. Although in this study
the samples distribution was not entirely consistent with the distribution of the
population, the sampling method used in this study did produce a representative pool
of internet users.
Furthermore, nearly all the respondents had more than one years internet
experience, and over 75 per cent of them had used the internet for more than three
years. Of the respondents, 77 per cent had experience of online shopping with the
online retailer that they frequently browsed, and 23 per cent had experience of
searching for products or service information online but had not made purchases. The
ten most often visited web sites were YAHOO! (30.1 per cent), Unimall (17.4 per cent),
PChome (9.9 per cent), ETMall (9.4 per cent), Books (6.7 per cent), P@YEASY (5.1 per
cent), HerMall (3.8 per cent), HappyBAG (2.9 per cent), ezTravel (2.5 per cent) and ERA
Ticket (2.4 per cent).

Non-recursive model
Recursive structural models assume that all causal effects are represented as
uni-directional and that there are no disturbance correlations between endogenous
OIR variables with direct effects between them. However, these assumptions are very
32,6 restrictive (Kline, 2005). For example, many real world causal processes are based on
cycles of mutual influence, that is, feedback. The presence of a feedback loop in a
structural model automatically makes it reciprocal. Non-recursive structural equation
models incorporate reciprocal feedback loops between variables. In the data-analytic
realm, a considerable volume of literature discusses the mechanics of the statistical
828 solution to such models (e.g. Berry, 1984; Kline, 2005; Martens and Haase, 2006; Wong
and Law, 1999). To test a reciprocal causal relationship between trust and perceived
risk and to confirm the proposed framework, this study applied a non-recursive
structural equation model, as recommended by Martens and Haase (2006) and Wong
and Law (1999).

Results
Consistent with the two-step approach advocated by Anderson and Gerbing (1988),
this study first developed the measurement model by conducting confirmatory factor
analysis (CFA). The structural equation model was then estimated for hypotheses
testing. Both the measurement model and the structural model were assessed by the
maximum likelihood method (Arbuckle, 2003) using AMOS 5.0. To evaluate the fit of
models, chi-square with degrees of freedom, the comparative fit index (CFI), the
goodness of fit index (GFI), the adjusted goodness of fit index (AGFI), the normed fit
index (NFI) and root mean square error of approximation (RMSEA) were assessed in
addition to the chi-square test. In general, model fit is considered to be adequate if GFI,
NFI and CFI are larger than 0.9, AGFI is larger than 0.8 and RMSEA is smaller than
0.08 (Bagozzi and Yi, 1988; Hair et al., 1998; Joreskog and Sorbom, 1994).

Measurement model
The measurement model yielded a chi-square value of 2,927.29 with 1,025 degrees of
freedom (P 0:000), indicating a general lack of fit. However, the chi-square test is
sensitive to sample sizes, especially where the size exceeds 200 respondents (Hair et al.,
1998). As an alternative, we used the ratio of the chi-square to the degrees of freedom.
This approach obtained a value of 2.856, which falls within the suggested value of 5 or
below (Bagozzi and Yi, 1988; Carmines and Mclver, 1981). In addition, with the
exception of the GFI, the other indices satisfied the recommended values (GFI 0:838,
NFI 0:903, CFI 0:914, AGFI 0:814, and RMSEA 0:054). The GFI at 0.838 was
slightly below the 0.9 benchmark, but it exceeded the recommended value of 0.80
suggested by Etezadi-Amoli and Farhoomand (1996). Therefore, there was a
reasonable overall fit between the model and the observed data.
Furthermore, each construct was evaluated separately by examining the indicator
loading for statistical significance and assessing the constructs reliability and
variance extracted (Hair et al., 1998). All of the loadings of the items on their latent
constructs were statistically significant (t-values . 2). As shown in Table I, the scale
composite reliabilities (rc) ranged from 0.71 to 0.93, exceeding the acceptable level of
0.70 (Hair et al., 1998). The average variance extracted for each construct ranged from
0.61 to 0.75, exceeding the acceptable level of 0.50 (Hair et al., 1998). The results
demonstrate convergent validity. To assess discriminant validity, the chi-square
difference test can be performed (Long, 1983). For every pair of constructs, two CFA
models are compared one where the correlation between the pair is constrained to
 Online store
Construct Cronbachs a Composite reliability Average variance extracted
environment
Technical adequacy 0.85 0.86 0.51 cues
Content quality 0.90 0.91 0.62
Special content 0.86 0.71 0.55
Appearance 0.90 0.90 0.65
Web site awareness 0.73 0.76 0.52 829
Web site image 0.80 0.80 0.57
Trust 0.92 0.92 0.66
Perceived risk 0.93 0.93 0.68
Purchase intention 0.88 0.88 0.71
Trust propensity 0.92 0.89 0.80
Risk propensity 0.76 0.76 0.60 Table I.
Measurement model fit
Notes: Criteria: Cronbachs a .0.70, composite reliability (rc) . 0.70, average variance extracted indices for reliability and
(AVE) . 0.50 (Hair et al., 1998) convergent validity

equal 1 and the other where it is allowed to vary. A significant chi-square difference
indicates discriminant validity between the pair of constructs (Bagozzi and Yi, 1988).
The results show that the chi-square differences ranged between 6.3 and 488.62
(p , 0:01), demonstrating adequate discriminant validity for all scales.
To confirm the existence of multi-dimensionality in the web site quality and web
site brand, this study performed CFA by comparing alternative measurement models
of both. Following the method used by Dabholkar et al. (1996) and Casalo et al. (2007),
this study compared a second order model in which various dimensions measured the
construct under consideration, with a first order model in which all the items weighted
a single factor (Steenkamp and Van Trijp, 1991). As shown in Table II, the results
reveal that the second order factor model of web site quality demonstrated acceptable
fit and performed better than the first order model in terms of all fitness measures. For
web site brand, the second order factor model and the first order factor model
demonstrated acceptable fit and performed very similarly for all fitness measures. We
thus conclude that the second-order factor model is suitable for modelling perceived
web site quality and web site brand.

Structural model
The structural model specified the trust propensity (j1), web site quality (j2), web site
awareness (j3) and risk propensity (j4) as the exogenous constructs, while trust (h1),
perceived risk (h2) and purchase intention (h3) were adopted as endogenous constructs.
Note that, as suggested by Kline (2005), the construct-level error terms of trust and

Construct Model x2 df GFI AGFI NFI CFI RMSEA

Web site quality Second order model 761.44 205 0.904 0.881 0.916 0.937 0.066
First order model 1,655.05 209 0.769 0.720 0.818 0.837 0.105
Web site brand Second order model 56.70 8 0.970 0.922 0.963 0.968 0.090
First order model 60.57 9 0.968 0.926 0.960 0.966 0.096
Table II.
Notes: Goodness-of-fit criteria: x 2/df: 1 , 5, GFI . 0.90, AGFI . 0.80, NFI . 0.90, CFI . 0.90, Comparison of
RMSEA , 0.08 (Bagozzi and Yi, 1988; Hair et al., 1998; Joreskog and Sorbom, 1994) measurement models
OIR perceived risk were permitted to correlate with each other in the model specification. In
32,6 addition, in the literature on non-recursive structural equation modelling, variables
added to a model to ensure that it is estimable are called instrumental variables
(Martens and Haase, 2006). In Figure 2, the exogenous variables of trust propensity and
risk propensity are instrumental variables, and their presence ensures that the model
can be estimated.
830 The structural model yielded a chi-square value of 1,000.54 with 282 degrees of
freedom (P 0:000). The ratio of the chi-square to the degrees of freedom was 3.548,
which was smaller than the recommended level of 5 (Bagozzi and Yi, 1988; Carmines
and Mclver, 1981). A comparison of all fit indices, with their corresponding
recommended values, indicated a good model fit (GFI 0:883, AGFI 0:854,
CFI 0:947, NFI 0:928, RMSEA 0:064), although the GFI value of 0.883 was at a
marginal acceptance level (Etezadi-Amoli and Farhoomand, 1996). The non-recursive
model thus satisfies the conditions of both order and rank (Kline, 2005; Martens and
Haase, 2006). Table III and Figure 2 show the structural model estimates, where the
estimate parameters are standardised path coefficients and where all path coefficients,
except the path of web site quality to perceived risk, are significant at the 95 per cent
level.
As reported in Table III, the analytical results indicate that web site quality is
related positively to trust, thus H1 is supported. Unexpectedly, web site quality does
not influence consumers perceived risk, thus H2 is not supported. Furthermore, the
brand of the retailers web site is positively related to consumers trust and negatively
related to perceived risk, providing support for H3 and H4. Also as hypothesised, the
relationships between consumers trust and perceived risk are non-recursive.

Figure 2.
Results of structural
equation model analysis
 Online store
Hypothesised relationship Sign Parameter Estimate t-value Conclusion
environment
H1 Web site quality ! trust l12 0.247 4.55 Supported cues
H2 Web site quality ! perceived risk 2 l22 0.029 0.29 Not supported
H3 Web site brand ! trust l13 0.139 2.40 Supported
H4 Web site brand ! perceived risk 2 l23 (0.325 (2.28 Supported
H5a Trust ! perceived risk 2 b21 20.297 2 2.15 Supported 831
H5b Perceived risk ! trust 2 b12 20.499 2 2.76 Supported
H6 Trust ! purchase intention b31 0.622 14.93 Supported
H7 Perceived risk ! purchase intention 2 b32 20.173 2 4.71 Supported
Control Trust propensity ! trust l11 0.392 6.73
variable Risk propensity ! perceived risk 2 l24 20.296 2 2.85
Trust ! R 2 0.69
Perceived risk ! R 2 0.21
Purchase intention ! R 2 0.50
Table III.
Notes: Standardised estimates are shown. R 2: Square multiple correlations; t-value is significant at Structural model
p , 0.05 while the critical ratios value exceeds 1.96 estimates

Consumers trust towards a specific online retailer significantly and negatively affects
perceived risk towards the online retailer. Similarly, perceived risk significantly
negatively affects consumers trust. Thus, H5a and H5b are supported. In terms of
purchase intention, consumers trust positively affects purchase intention towards a
specific web site, while perceived risk negatively affects purchase intention, providing
support for H6 and H7. Altogether, predictors accounted for 69 per cent of the
variation in trust, 21 per cent of the variation in perceived risk and 50 per cent of the
variation in purchase intention.

Discussion
The results of this study provide support for the research framework presented in
Figure 1 and for the hypotheses regarding the directional linkages among the model
variables. The analytical results demonstrate that perceptions of web site quality and
web site brand affect consumers trust and perceived risk, and in turn, consumers
purchase intention towards a specific online retailer. Additionally, the results reveal
that the relationships between trust and perceived risk are non-recursive. The
relationships among the research constructs are discussed below.
First, this study shows that web site quality and web site brand contribute to
consumers trust and perceived risk, and in turn, purchase intention. The total effects
of web site quality and web site brand on customers trust, perceived risk and purchase
intention are (0.290, 2 0.086 and 0.195) and (0.353, 2 0.430 and 0.294), respectively (see
Table IV). These results demonstrate that web site brand exerts a stronger effect on
consumers decision making than web site quality, implying that a well-established
web site brand may prove to be valuable in inspiring high confidence in consumers and
increasing their purchase intention. For example, the relationships among web site
brand, consumers trust, perceived risk and purchase intention indicate that consumers
who are familiar with and recognise the brand of the online retailer are more likely to
trust, less likely to perceive risk, and in turn more likely to purchase from the online
retailer.
OIR
Predictor/Dependent Web site quality Web site brand Trust Perceived risk
32,6
Direct effects
Trust 0.247 0.139
Perceived risk n.s. 2 0.325 2 0.297 20.499
Purchase intention 0.622 20.173
832 Indirect effects
Trust 0.043 0.215 0.174 20.087
Perceived risk 2 0.086 2 0.105 2 0.052 0.174
Purchase intention 0.195 0.294 0.169 20.395
Total effects
Trust 0.290 0.354 0.174 20.586
Perceived risk 2 0.086 2 0.430 2 0.349 0.174
Table IV. Purchase intention 0.195 0.294 0.790 20.567
Direct, indirect and total
effects estimates Notes: Standardised estimates are shown. All non-zero effects are significant at p , 0:05

These findings are similar to those of Ennew et al. (2005). In a web site traffic study,
they concluded that web site brand reputation may be an important factor in
explaining higher predicted numbers of visitors. In addition, according to a study by
Cheskin Research, brand is at the top of the list of six marketplace fundamentals for
building and maintaining trust on the internet (the others are navigability, fulfilment,
presentation, technology and seals of approval) (Ibeh et al., 2005), consistent with our
findings.
Furthermore, although web site quality has no significant direct effect on perceived
risk, this study reveals that web site quality appears to indirectly affect perceived risk
through customers trust towards the online retailer (see Table IV). These findings
demonstrate that web site quality presents tangible cues that can be deliberately used
to build consumer trust in a retailers ability, integrity and benevolence, and in turn,
reduce customer-perceived uncertainty and risk towards the online retailer.
Finally, the results also show that consumers trust and perceived risk towards a
specific online retailer mediate the web site quality and web site brand with regards to
purchase intention. Note that the reciprocal path coefficient between consumers trust
and perceived risk are of the same sign (2 0.30 and 2 0.50). The analysis suggests that
the non-recursive model is tenable in the sense that the reciprocal paths are not equal
and that both are significantly different from 0. The non-recursive analysis of these
data yields a fairly unambiguous interpretation there is evidence within these data
that the relationships between consumers trust and perceived risk may well be
reciprocal in nature. As consumers trust increases because of decreasing perceived
risk, further decreases in perceived risk will follow. This cycle would be repeated
throughout the feedback loop until damping down. The analysis results suggest that
the causal influence between consumers trust and perceived risk is bi-directional and
reciprocal, consistent with the view of Mitchell (1999). This finding also confirms the
research of Rousseau et al. (1998), which suggested that trust has a dynamic, rather
than static nature. Trust can change over time (developing, building and declining) and
hence has different levels under different conditions the perceived level of risk and
interdependence between two parties. This study shows that the level of trust changes
with the variations of perceived risk, consistent with the view of Rousseau et al. (1998).
Conclusions and implications Online store
The results of this study shed light on some important issues related to online environment
environment cues and consumers purchase decisions that have not been addressed by
previous studies. First, although previous research on online store environment cues cues
has focused narrowly on web site quality (Eroglu et al., 2001, 2003; Ethier et al., 2006),
this study found that web site brand is a critical cue that affects consumer purchase
intention. Notably, this study reveals that web site brand is a more influential factor 833
than web site quality in online purchase decision-marking, implying that managing the
brand of the web site is more important than emphasising the interface quality in the
context of online shopping. This finding is particularly important for managers of
online retailers as they decide how to allocate resources. However, managing the web
site brand is much more difficult, time-consuming and expensive than improving the
functionality of the web site (Yoon, 2002). Online retailers need to search for e-brand
building strategies that might assist in creating distinctiveness and engaging
customers.
This study suggests that online retailers should consider focusing their marketing
strategies more on establishing the awareness and image of the web site. For example,
retailers could be either exclusively online (e.g. Amazon) or have both an online and
brick-and-mortar presence (e.g. Wal-Mart). A retailer that has both online and offline
presence has some advantage, since the awareness and image built for the
brick-and-mortar business can be transferred to the online store. Thus, managers
should leverage their company name to generate awareness of their online presence
since customers with cognitive and affective bonds (e.g. trust) with the retailer will
tend to make a purchase decision even though the web site is not superior to that of
other retailers.
Furthermore, previous studies that tested the stimulus-organism-response (S-O-R)
framework in a retail store environment usually adopted the
pleasure-arousal-dominance (PAD) emotional experience as the intervening
organismic state. However, Eroglu et al. (2003) concluded that the PAD emotion
scheme does not capture much of the effect of site atmosphere. They thus
recommended further study using an emotion scheme that is more comprehensive in
nature (such as interest, surprise, anger/frustration and disgust). This study argues
that trust and perceived risk can be viewed as cognitive and affective intermediary
states that fluctuate with online environment cues.
The analytical results show the total effects of web site quality and web site brand
on consumers trust (0.290, 0.353) and perceived risk (2 0.086, 2 0.430). These results
demonstrate that consumer trust and perceived risk could well capture the effect of
online environment cues, although web site quality appears to indirectly affect
perceived risk. Therefore, this study suggests that future research could consider using
trust and perceived risk as organism variables in the stimulus-organism-response
research framework.
Finally, this study represents the first empirical examination of the reciprocal causal
links between the constructs of consumers trust and perceived risk, with most
previous research having considered the relationships between them to be
uni-directional. Although non-recursive structural equation models are far less well
known and far less frequently used than recursive mediated models, non-recursive
structural equation models represent processes that appear frequently in psychological
OIR theory (Martens and Haase, 2006). This study demonstrates that the causal influence
32,6 between consumers trust and perceived risk is bi-directional and reciprocal, which
contributes to providing empirical support for the hypothesised non-recursive relation
of trust and perceived risk in the context of an online environment.

Limitations and future research

834 As with any research, care should be taken when generalising the results of this study.
First, as was mentioned in the literature review section, web site quality is a
multi-faceted concept. This study adopted the four dimensions (technical adequacy,
content quality, specific content and appearance.) of web site quality developed by
Aladwani and Palvia (2002). However, other components of web site quality may yield
different results. Thus, further studies using other dimensional measures may be
needed to verify the research model.
Second, while this study focused on users perceptions of web sites based on
browsing experience, as the browsing took place prior to the data collection, the effect
of timing on the changes in their perception of the importance of various web quality
factors was not discovered. In addition, various factors such as internet experience, the
type of web site (e.g. music CDs, books, travel, computers, apparel and cosmetics) and
incentive programmes might also complicate the users perceptions of web site quality,
awareness, image, trust and risk, but these factors were considered beyond the scope of
this study. Further studies are necessary to explore their impact on consumers online
purchase intention.
Finally, the survey was conducted using web-based forms and employed a
non-random, convenient sample. Gathering a larger sample using an alternative
survey modality and random sampling methods would be costly. The online survey
method was appropriate for collecting data from participants with internet experience
who were also free of geographical constraints. Although the sampling method used in
this study was adequately suited to its purpose, the generalisability could be enhanced
if future research systematically sampled from a more diverse population.

References
Aaker, D.A. (1991), Managing Brand Equity: Capitalizing on the Value of Brand Name, Free
Press, New York, NY.
Aaker, D.A. (1996), Measuring brand equity across products and markets, California
Management Review, Vol. 38 No. 3, pp. 102-20.
Ahn, T., Ryu, S. and Han, I. (2007), The impact of web quality and playfulness on user
acceptance of online retailing, Information & Management, Vol. 44 No. 3, pp. 263-75.
Aladwani, A.M. (2006), An empirical test of the link between web site quality and forward
enterprise integration with web consumers, Business Process Management Journal,
Vol. 12 No. 2, pp. 178-90.
Aladwani, A.M. and Palvia, P.C. (2002), Developing and validating an instrument for measuring
user-perceived web quality, Information & Management, Vol. 39 No. 6, pp. 467-76.
Albert, C. (2004), The impact of switching costs on customer loyalty: a study among corporate
customers of mobile telephony, Journal of Targeting, Measurement and Analysis for
Marketing, Vol. 12 No. 3, p. 256.
Anderson, J.C. and Gerbing, D.W. (1988), Structural equation modeling in practice: a review and
recommended two-step approach, Psychological Bulletin, Vol. 103 No. 3, pp. 411-23.
Arbuckle, J.L. (2003), Amos 5 [Computer Software], SmallWaters, Chicago, IL. Online store
Arcand, M., Nantel, J., Arles-Dufour, M. and Vincent, A. (2007), The impact of reading a web environment
sites privacy statement on perceived control over privacy and perceived trust, Online
Information Review, Vol. 31 No. 5, pp. 661-81. cues
Bagozzi, R.P. and Yi, Y. (1988), On the evaluation of structural equation models, Journal of the
Academy of Marketing Science, Vol. 16 No. 1, pp. 74-94.
Baker, J. (1986), The role of environment in marketing services: the consumer perspective, in 835
Czepiel, J.A., Congram, C.A. and Shanahan, J. (Eds), The Services Challenge: Integrating for
Competitive Advantage, American Marketing Association, Chicago, IL, pp. 79-84.
Bauer, H.H., Falk, T. and Hammerschmidt, M. (2006), eTransQual: a transaction process-based
approach for capturing service quality in online shopping, Journal of Business Research,
Vol. 59 No. 6, pp. 875-86.
Berry, L.L. (2000), Cultivating service brand equity, Journal of the Academy of Marketing
Science, Vol. 28 No. 1, pp. 128-37.
Berry, W.L. (1984), Non-recursive Causal Models, Sage, Thousand Oaks, CA.
Bhattacherjee, A. (2002), Individual trust in online firms: scale development and initial test,
Journal of Management Information Systems, Vol. 19 No. 1, pp. 211-41.
Biel, A.L. (1992), How brand image drives brand equity, Journal of Advertising Research, Vol. 32
No. 6, pp. 6-12.
Bitner, M.J. (1992), Servicescapes: the impact of physical surroundings on customers and
employees, Journal of Marketing, Vol. 56, April, pp. 57-71.
Blackwell, R.D., Miniard, P.W. and Engel, J.F. (2001), Consumer Behavior, Harcourt College
Publishers, Troy, MO.
Boksberger, P.E., Bieger, T. and Laesser, C. (2007), Multidimensional analysis of perceived risk
in commercial air travel, Journal of Air Transport Management, Vol. 13 No. 2, pp. 90-6.
Carmines, E.G. and Mclver, J.P. (1981), Analyzing models with unobserved variables: analysis of
covariance structure, in Bohrnstedt, G.W. and Borgatta, E.F. (Eds), Social Measurement:
Current Issues, Sage, Newbury Park, CA.
Casalo, L.V., Flavian, C. and Guinalu, M. (2007), The role of security, privacy, usability and
reputation in the development of online banking, Online Information Review, Vol. 31 No. 5,
pp. 583-603.
Chang, H.H. (2008), Intelligent agents technology characteristics applied to online auctions
task: a combined model of TTF and TAM, Technovation, Vol. 28 No. 9, pp. 564-79.
Chang, H.H. and Wang, I.C. (2008), An investigation of user communication behavior in
computer mediated environments, Computers in Human Behavior, Vol. 24 No. 5,
pp. 2336-56.
Chang, M.K., Cheung, W. and Lai, V.S. (2005), Literature derived reference models for the
adoption of online shopping, Information & Management, Vol. 42 No. 4, pp. 543-59.
Chau, P.Y.K., Hu, P.J.-H., Lee, B.L.P. and Au, A.K.K. (2007), Examining customers trust in online
vendors and their dropout decisions: an empirical study, Electronic Commerce Research
and Applications, Vol. 6 No. 2, pp. 171-82.
Chen, S.C. and Dhillon, G.S. (2003), Interpreting dimensions of consumer trust in e-commerce,
Information Technology and Management, Vol. 4 Nos 2-3, pp. 303-18.
Chen, Y.-H. and Barnes, S. (2007), Initial trust and online buyer behaviour, Journal of Industrial
Management & Data Systems, Vol. 107 No. 1, pp. 21-36.
Cho, J. (2004), Likelihood to abort an online transaction: influences from cognitive evaluations,
attitudes, and behavioral variables, Information & Management, Vol. 41 No. 7, pp. 827-38.
OIR Cho, J. and Lee, J. (2006), An integrated model of risk and risk-reducing strategies, Journal of
Business Research, Vol. 59 No. 1, pp. 112-20.
32,6 Churchill, G.A.J. (1995), Marketing Research: Methodological Foundations, 6th ed., The Dryden
Press, New York, NY.
Corbitt, B.J., Thanasankit, T. and Yi, H. (2003), Trust and e-commerce: a study of consumer
perceptions, Electronic Commerce Research and Applications, Vol. 2 No. 3, pp. 203-15.
836 Corritore, C.L., Kracher, B. and Wiedenbeck, S. (2003), On-line trust: concepts, evolving themes,
a model, International Journal of Human-Computer Studies, Vol. 58 No. 6, pp. 737-58.
Dabholkar, P.A., Thorpe, D.I. and Rentz, J.O. (1996), A measure of service quality for retail
stores: scale development and validation, Journal of the Academy of Marketing Science,
Vol. 24, Winter, pp. 3-16.
Davis, D.F., Golicic, S.L. and Marquardt, A.J. (2007), Branding a B2B service: does a brand
differentiate a logistics service provider?, Industrial Marketing Management, Vol. 61
No. 8.
Davis, L., Wang, S. and Lindridge, A. (2007), Culture influences on emotional responses to
on-line store atmospheric cues, Journal of Business Research, Vol. 37 No. 2.
de Ruyter, K., Wetzels, M. and Kleijnen, M. (2001), Customer adoption of e-service: an
experimental study, International Journal of Service Industry Management, Vol. 12 No. 2,
pp. 184-207.
Ennew, C., Lockett, A., Blackman, I. and Holland, C.P. (2005), Competition in internet retail
markets: the impact of links on web site traffic, Long Range Planning, Vol. 38 No. 4,
pp. 359-72.
Eroglu, S.A., Machleit, K.A. and Davis, L.M. (2001), Atmospheric qualities of online retailing: a
conceptual model and implications, Journal of Business Research, Vol. 54 No. 2, pp. 177-84.
Eroglu, S.A., Machleit, K.A. and Davis, L.M. (2003), Empirical testing of a model of online store
atmospherics and shopper responses, Psychology and Marketing, Vol. 20 No. 2, pp. 139-50.
Etezadi-Amoli, J. and Farhoomand, A.F. (1996), A structural model of end user computing
satisfaction and user performance, Information & Management, Vol. 30 No. 2, pp. 65-73.
Ethier, J., Hadaya, P., Talbot, J. and Cadieux, J. (2006), B2C web site quality and emotions during
online shopping episodes: an empirical study, Information & Management, Vol. 43 No. 5,
pp. 627-39.
Fam, K.S., Foscht, T. and Collins, R.D. (2004), Trust and the online relationships an
exploratory study from New Zealand, Tourism Management, Vol. 25 No. 2, pp. 195-207.
Forsythe, S.M. and Shi, B. (2003), Consumer patronage and risk perceptions in internet
shopping, Journal of Business Research, Vol. 56 No. 11, pp. 867-75.
Gefen, D. (2000), E-commerce: the role of familiarity and trust, Omega, Vol. 28 No. 6, pp. 725-37.
Gefen, D. and Straub, D.W. (2004), Consumer trust in B2C e-commerce and the importance of
social presence: experiments in e-products and e-services, Omega, Vol. 32 No. 6,
pp. 407-24.
Grabner-Krauter, S. and Kaluscha, E.A. (2003), Empirical research in on-line trust: a review and
critical assessment, International Journal of Human-Computer Studies, Vol. 58 No. 6,
pp. 783-812.
Grace, D. and OCass, A. (2005), Service branding: consumer verdicts on service brands, Journal
of Retailing and Consumer Services, Vol. 12 No. 2, pp. 125-39.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis, 5th
ed., Prentice-Hall, Englewood Cliffs, NJ.
Harridge-March, S. (2006), Can the building of trust overcome consumer perceived risk online?,
Marketing Intelligence & Planning, Vol. 24 No. 7, pp. 746-61.
Hashim, N.H. and Murphy, J. (2007), Branding on the web: evolving domain name usage among Online store
Malaysian hotels, Tourism Management, Vol. 28 No. 2, pp. 621-4.
Hoffman, D.L., Novak, T.P. and Peralta, M. (1999), Building consumer trust online,
environment
Communications of the ACM, Vol. 42 No. 4, pp. 80-5. cues
Hwang, Y. and Kim, D.J. (2007), Customer self-service systems: the effects of perceived web
quality with service contents on enjoyment, anxiety, and e-trust, Decision Support
Systems, Vol. 43 No. 3, pp. 746-60. 837
Ibeh, K.I.N., Ying, L. and Dinnie, K. (2005), E-branding strategies of internet companies: some
preliminary insights from the UK, Journal of Brand Management, Vol. 12 No. 5,
pp. 355-73.
Joreskog, K. and Sorbom, D. (1994), Structural Equation Modeling with the SIMPLIS Command
Language, Scientific Software International, Chicago, IL.
Jari, S. and Heikki, K. (2007), A conceptual model of trust in the online environment, Online
Information Review, Vol. 31 No. 5, pp. 604-21.
Johnson, D. and Grayson, K. (2005), Cognitive and affective trust in service relationships,
Journal of Business Research, Vol. 58 No. 4, pp. 500-7.
Keller, K.L. (1993), Conceptualizing, measuring, and managing customer-based brand equity,
Journal of Marketing, Vol. 57 No. 1, pp. 1-22.
Keller, K.L. (2003), Strategic Brand Management: Building, Measuring, and Managing Brand,
2nd ed., Prentice-Hall, Englewood Cliffs, NJ.
Kim, D.J., Ferrin, D.L. and Rao, H.R. (2007), A trust-based consumer decision-making model in
electronic commerce: the role of trust, perceived risk, and their antecedents, Decision
Support Systems.
Kline, R.B. (2005), Principles and Practice of Structural Equation Modeling, 2nd ed., Guildford
Publications, Inc., New York, NY.
Koufaris, M. and Hampton-Sosa, W. (2004), The development of initial trust in an online
company by new customers, Information & Management, Vol. 41 No. 3, pp. 377-97.
Lee, K.S. and Lee, H.S. (2007), Factors influencing the adoption behavior of mobile banking: a
South Korean perspective, Journal of Internet Banking and Commerce, Vol. 12 No. 2,
pp. 1-9.
Lewis, D. and Weigert, A. (1985), Trust as a social reality, Social Forces, Vol. 63 No. 4,
pp. 967-85.
Liao, C., Palvia, P. and Lin, H.-N. (2006), The roles of habit and web site quality in e-commerce,
International Journal of Information Management, Vol. 26 No. 6, pp. 469-83.
Lim, N. (2003), Consumers perceived risk: sources versus consequences, Electronic Commerce
Research and Applications, Vol. 2 No. 3, pp. 216-28.
Limayem, M. and Khalifa, M. (2000), What makes consumers buy from the internet? A
longitudinal study of online shopping, IEEE Transactions on Systems, Man &
Cybernetics: Part A, Vol. 30 No. 4, p. 421.
Lin, W.-B. (2007), Investigation on the model of consumers perceived risk integrated
viewpoint, Expert Systems with Applications.
Long, J. (1983), Confirmatory Factor Analysis, Sage Publications, Beverly Hills, CA.
McKnight, D.H., Kacmar, C.J. and Choudhury, V. (2004), Dispositional trust and distrust
distinctions in predicting high-and low-risk internet expert advice site perceptions,
e-Service Journal, Vol. 3 No. 2, pp. 35-58.
McKnight, H.D., Choudhury, V. and Kacmar, C. (2002), The impact of initial consumer trust on
intentions to transact with a web site: a trust building model, The Journal of Strategic
Information Systems, Vol. 11 Nos 3-4, pp. 297-323.
OIR Martens, M.P. and Haase, R.F. (2006), Advanced applications of structural equation modeling in
counseling psychology research, The Counseling Psychologist, Vol. 34 No. 6, pp. 878-911.
32,6 Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995), An integrative model of organizational
trust, Academy of Management Review, Vol. 20 No. 3, pp. 709-34.
Mitchell, V.-W. (1999), Consumer perceived risk: conceptualizations and models, European
Journal of Marketing, Vol. 33 Nos 1/2, pp. 163-95.
838 Mitchell, V.W. (2001), Re-conceptualizing consumer store image processing using perceived
risk, Journal of Business Research, Vol. 54 No. 2, pp. 167-72.
Mullins, J.W. and Forlani, D. (2005), Missing the boat or sinking the boat: a study of new venture
decision making, Journal of Business Venturing, Vol. 20 No. 1, pp. 47-69.
Napier, H.A., Judd, P.J., Rivers, O.N. and Wagner, S.W. (2001), Creating a Winning E-business,
Thomson Learning, Canada.
Negash, S., Ryan, T. and Igbaria, M. (2003), Quality and effectiveness in web-based customer
support systems, Information & Management, Vol. 40 No. 8, pp. 757-68.
Pavlou, P.A. (2003), Consumer acceptance of electronic commerce: integrating trust and risk
with the technology acceptance model, International Journal of Electronic Commerce,
Vol. 7 No. 3, pp. 101-34.
Ranganathan, C. and Ganapathy, S. (2002), Key dimensions of business-to-consumer web sites,
Information & Management, Vol. 39 No. 6, pp. 457-65.
Reichheld, F.F. and Schefter, P. (2000), E-loyalty: your secret weapon on the web, Harvard
Business Review, Vol. 78 No. 4, pp. 105-13.
Riegelsberger, J., Sasse, M.A. and McCarthy, J.D. (2003), The researchers dilemma: evaluating
trust in computer-mediated communication, International Journal of Human-Computer
Studies, Vol. 58 No. 6, pp. 759-81.
Romaniuk, J. and Sharp, B. (2003), Measuring brand perceptions: testing quantity and quality,
Journal of Targeting, Measurement and Analysis for Marketing, Vol. 11 No. 3, pp. 218-29.
Rousseau, D., Sitkin, S., Burt, R. and Camerer, R. (1998), Not so different after all: a cross
discipline view of trust, Academy of Management Review, Vol. 23 No. 3, pp. 393-404.
Rowley, J. (2004), Online branding, Online Information Review, Vol. 28 No. 2, pp. 131-8.
Sherman, E., Mathur, A. and Smith, R.B. (1997), Store environment and consumer purchase
behavior: mediating role of consumer emotions, Psychology and Marketing, Vol. 14 No. 4,
pp. 361-78.
Sitkin, S.B. and Pablo, A.L. (1992), Reconceptualizing the determinants of risk behavior,
Academy of Management Review, Vol. 17 No. 1, pp. 9-38.
Smith, D.N. and Sivakumar, K. (2004), Flow and internet shopping behavior: a conceptual model
and research propositions, Journal of Business Research, Vol. 57 No. 10, pp. 1199-208.
Steenkamp, J.B. and Van Trijp, H.C.M. (1991), The use of LISREL in validating marketing
constructs, International Journal of Research in Marketing, Vol. 8, pp. 283-99.
Stewart, K.J. (1999), Transference as a means of building trust in world wide web sites,
Proceedings of the 20th International Conference on Information Systems, Association for
Information Systems, Atlanta, GA, pp. 459-64.
Teo, T.S.H. and Liu, J. (2007), Consumer trust in e-commerce in the United States, Singapore
and China, Omega, Vol. 35 No. 1, pp. 22-38.
TWNIC (2008), Taiwan Network Information Center (TWNIC), available at: www.twnic.net/
index2.php
Walczuch, R. and Lundgren, H. (2004), Psychological antecedents of institution-based consumer
trust in e-retailing, Information & Management, Vol. 42 No. 1, pp. 159-77.
Wang, Y.D. and Emurian, H.H. (2005), An overview of online trust: concepts, elements, and Online store
implications, Computers in Human Behavior, Vol. 21 No. 1, pp. 105-25.
Wilde, S.J., Kelly, S.J. and Scott, D. (2004), An exploratory investigation into e-tail image
environment
attributes important to repeat, internet savvy customers, Journal of Retailing and cues
Consumer Services, Vol. 11 No. 3, pp. 131-9.
Wolfinbarger, M. and Gilly, M.C. (2003), eTailQ: dimensionalizing, measuring and predicting
etail quality, Journal of Retailing, Vol. 79 No. 3, pp. 183-98. 839
Wong, C.-S. and Law, K.S. (1999), Testing reciprocal relations by non-recursive structural
equation models using cross-sectional data, Organizational Research Methods, Vol. 2
No. 1, pp. 69-87.
Wu, J.-H. and Wang, S.-C. (2005), What drives mobile commerce? An empirical evaluation of the
revised technology acceptance model, Information & Management, Vol. 42 No. 5,
pp. 719-29.
Yoon, S.-J. (2002), The antecedents and consequences of trust in online-purchase decisions,
Journal of Interactive Marketing, Vol. 16 No. 2, pp. 47-63.

Appendix. Instrument items

(Note: X was replaced by the name of the retailers web site in the questionnaire.)
Perceived web site quality (Aladwani and Palvia, 2002)

Technical adequacy
1. Xs web site looks easy to navigate through.
2. Xs web site has adequate search facilities.
3. Xs web site is always up and available.
4. Xs web site has valid links (hyperlinks).
5. Xs web site has many interactive features.
6. Xs web site is easy to access.

Content quality
7. The content of Xs web site is useful.
8. The content of Xs web site is complete.
9. The content of Xs web site is clear.
10. The content of Xs web site is current.
11. The content of Xs web site is concise.
12. The content of Xs web site is accurate.

Special content
13. In Xs web site, one can find contact information (e.g. email addresses, phone numbers,
etc.).
OIR 14. In Xs web site, one can find the firms general information (e.g. goals, owners).
32,6 15. In Xs web site, one can find details about products and/or services.
16. In Xs web site, one can find information related to customers policies (e.g. privacy and
dispute details).
17. In Xs web site, one can find information related to customer service.
840
Appearance
18. Xs web site looks attractive.
19. Xs web site looks organised.
20. Xs web site uses fonts properly.
21. Xs web site uses colours properly.
22. Xs web site uses multimedia features properly.
Web site awareness (Aaker, 1996; Davis et al., 2007)
1. The name of Xs web site is well-known in the online retailer industry.
2. Xs web site is a comparatively recallable web site to me.
3. I can recognise Xs web site among other competing web sites.
Web site image (Aaker, 1996; Davis et al., 2007)
1. Xs web site is known as one that takes good care of their customers.
2. In comparison to other retailers web sites, Xs web site is known to consistently deliver
very high quality.
3. Xs web site is made by an organisation I would trust.
Trust (Gefen and Straub, 2004; Hwang and Kim, 2007)
1. Xs web site is competent.
2. Xs web site knows how to provide excellent service.
3. Promises made by Xs web site are likely to be reliable.
4. I expect that Xs web site will keep promises they make.
5. I expect that Xs web site has good intentions towards me.
6. I expect that Xs web sites intentions are benevolent.
Perceived risk (Corbitt et al., 2003)
1. I believe that online purchases from Xs web site are risky because the
products/services delivered may fail to meet my expectations.
2. I believe that online purchases from Xs web site are risky because the
products/services delivered may be inferior.
 3. I believe that online purchases from Xs web site are risky because they may lead to Online store
financial loss for me.
environment
4. I believe that online purchases from Xs web site are risky because they may cause cues
others to think less highly of me.
5. I believe that online purchases from Xs web site are risky because the
products/services delivered may fail to fit well with my personal image or self-concept. 841
6. I believe that online purchases from Xs web site are risky because they may lead to a
time loss for me.
Purchase intention (Pavlou, 2003)
1. I intend to use Xs web site to conduct product purchases.
2. I expect to purchase through Xs web site in the future.
3. It is likely that I will transact with Xs web site in the near future.
Trust propensity (Chen and Barnes, 2007; Koufaris and Hampton-Sosa, 2004)
1. It is easy for me to trust a person/thing.
2. I tend to trust a person/thing, even though I have little knowledge of them/it.
3. Trusting someone or something is not difficult.
Risk propensity (Cho and Lee, 2006)
1. I am willing to take substantial risks to do online shopping.
2. I am willing to accept some risk of losing money if online shopping is likely to involve
an insignificant amount of risk.

Corresponding author
Hsin Hsin Chang can be contacted at: easyhhc@mail.ncku.edu.tw

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints