Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
This paper investigates how consumers perceive and adopt or resist online shopping from
the expectation confirmation model (ECM), technology acceptance model (TAM), theory of
planned behavior (TPB), and perceived risk to identify the determinants which influence
consumers’ online shopping decision processes. The proposed model is then validated
empirically using a longitudinal method with a two-stage survey. The results demonstrate that
satisfaction has the most significant effect on consumers’ continuance intention, followed by
perceived usefulness as a significant but weaker predictor, and also that the initial intention to
use online shopping is adversely affected mainly by perceived risk. This study is significant
for two reasons. First, it synthesizes the expectation confirmation model, technology
help bride the existing gap among resistance, initial acceptance and continuance intention to
1
1. Introduction
With the rapid growth of the Internet, online shopping has become one of the most
profitable e-commerces on the Internet over the last decade (Vijayasarathy 2004). However,
online shopping is still confronted with many obstacles. First, considerable Internet users are
still reluctant to use online shopping (Forsythe and Shi, 2003; Soopramanien and Robertson,
using online shopping after initially accepting it) frequently occurs (Bhattacherjee 2001).
For the former issue, several authors (Forsythe and Shi, 2003; Hsu et al., 2006) have
attributed consumers’ reluctance to purchase online to apparent risks (e.g. credit card stolen
risk, identity theft risk) but these risks and their potential impact on consumers’ continuance
online purchase behavior are unclear. For the latter issue, prior studies have given more
attention to examining factors that drive users to “initially adopt online shopping” (i.e., use
online shopping for the first time), rather than the factors that influence uses to “continue to
use online shopping” after they have adopted it. Although initial acceptance of online
shopping is an important first step toward achieving success of online stores, the eventual
success of online stores still needs customers to repurchase things online. Therefore, with a
better understanding of the consumers’ online purchase decision processes from resistance
and initial adoption to continued intention to use online shopping, practitioners should be able
to make adjustments in their marketing plans that should help to provide them with a
In order to fill this knowledge gap, this study proposed a theoretical model which
synthesizes the expectation confirmation model (ECM), technology acceptance model (TAM),
theory of planned behavior (TPB), and perceived risk to understand consumers’ online
theoretical models and perspectives for the following three reasons. First, although previous
2
research has found ECM to be a robust model for continued IT adoption (Bhattacherjee,
2001), it employs only three variables to explain behavioral intention, namely satisfaction,
adopting IT will also be affected by other factors, such as the opinions of important
individuals (subjective norms) (Fishbein and Ajzen, 1975). Furthermore, even if users have a
strong intention to perform a behavior, they may feel that they lack the necessary resources
and skills (perceived behavioral control) (Ajzen, 1991), and the use of TPB addresses this
gap.
Second, while TPB captures the roles of individuals, organizational members, and social
influences on behavioral intention, it does not inform us what attitudinal beliefs would affect
a user’s attitude toward online shopping (Hsu et al., 2006). According to Taylor’s research
(Taylor and Todd, 1995a), TAM provides two attitudinal beliefs, namely perceived ease of
use and perceived usefulness as two major antecedents of attitude, which make up precedent
factors of attitude for TPB. Moreover, since each theory has distinct roots and is based on a
different set of antecedent variables, we contend that they independently provide a partial
when combined, these theories may collectively provide an improved and more
Third, customers have shown reluctance to complete online purchases, primarily due to
risk concerns (Gefen et al., 2003). This is because there is a lot at stake for consumers as they
entering into a business relationship with distant, faceless online purchases. While perceived
risk has been examined in the online shopping behavior, little has been known about its
influence on the consumers’ continuance intention to shop online, which may help
3
perceived risk to examine customer’s initial acceptance and continued usage of online
shopping.
While prior research has examined ECM, TAM, and TPB independently in explaining
information technology (IT) usage, to the best of our knowledge, no study has yet
theoretically combined these three models. The primary contributions of this study are its
examination of the integration of ECM, TAM, TPB and perceived risk in explaining initial
acceptance and continuance intention toward online shopping and an empirical evaluation of
which factors are critical to affecting this intention. The findings from this paper may
therefore help bridge the existing gap between initial acceptance and continuance streams of
online shopping research. The results of this work would be useful for online store
The rest of the paper is organized as follows. The next section presents our research
model and hypotheses, while Section 3 proposes the measurement method and scales. We
present the research results in Section 4, followed by discussion in Section 5. Finally, the
general (Anderson and Sullivan, 1993; Oliver, 1980, 1993). In order to understand users’
continued information technology (IT) usage behavior, Bhattacherjee (2001) adapt the ECT
to propose the expectation confirmation model (ECM), which emphasizes the importance of
4
already explained by confirmation variable, it does not include the performance variable.
The ECM posits that an individual’s intention to continue IT usage is dependent on three
variables: the user’s level of satisfaction with the IT; the extent of user’s confirmation of
Post-Adoption Expectations
(Perceived Usefulness)
There are five main hypotheses in the ECM. First, users’ satisfaction with IT has a positive
effect on their intention to continue using the IT. Studies in marketing have discovered that
the major reason for a consumer’s decision to repurchase products or patronize services is
their level of satisfaction (e.g. (Bearden and Teel, 1983; Oliver, 1993; Szymanski and Henard,
context and the continued usage of IT products/services, the ECM posits an equivalent
relationship in the latter context. In turn, user’s satisfaction with IT is determined by the
user’s confirmation of expectations and their perceived usefulness of IT (which is one type of
expected benefits through their usage experiences with the IT, and thus leads to a positive
5
paradigm, users’ post-usage perceived usefulness of IT has a positive effect on their
This relationship is supported by the adaptation level theory, which proposes that users
Davis (1989) first introduced the TAM as a theoretical extension of the theory of
reasoned action (TRA) (Fishbein and Ajzen, 1975) and found that it could better explain
user’s acceptance. TAM proposes that two particular beliefs, perceived usefulness and
perceived ease of use, are the primary drivers for technology acceptance. Perceived
usefulness is defined as ‘‘the degree to which a person believes that using a particular system
would enhance his/ her job performance’’, and perceived ease of use is defined as ‘‘the
degree to which a person believes that using a particular system would be free of physical and
mental effort’’ (Davis, 1989). Further, perceived usefulness and perceived ease of use both
affect a person’s attitude toward using the system, and consistent with TRA, these attitudes
toward using the system determine behavioral intentions, which in turn lead to actual system
use. TAM has been extensively applied to user acceptance research of various types of
technologies including e-mail, word processor, world wide web, enterprise resources
planning (ERP) systems, and e-commerce (Davis, 1989; Gefen and Straub, 1997; Lu et al.,
2009).
Both the theory of planned behavior (TPB) and TAM were developed based on the
theory of reasoned action (TRA) (Ajzen and Fishbein, 1980; Fishbein and Ajzen, 1975),
which argues that both behavioral attitude and subjective norm affect behavioral intention,
which in turn affects the actual behavior. TPB adds to TRA a third factor – perceived
behavioral control – that affects behavioral intention and actual behavior (Ajzen, 1991).
6
Many studies have replicated and investigated these three constructs and found that they are
valid in explaining individual intention to use various forms of IT (Liao, 1999; Venkatesh,
2000).
Since the 1960s, perceived risk theory has been used to explain consumers’ behavior.
Considerable research has examined the impact of risk on traditional consumer decision
making (Lin, 2008). Peter and Ryan (1976) defined perceived risk as a kind of subjective
expected loss, and Featherman and Savlou (Featherman and Pavlou, 2003) also defined
perceived risk as the possible loss when pursuing a desired result. Cunningham(Cunningham,
1967) noted that perceived risk consisted of the size of the potential loss (i.e. that which is at
stake) if the results of the act were not favorable and the individual’s subjective feelings of
certainty that the results will not be favorable. Most of scholars claimed that consumers’
perceived risk have been identified: financial, performance, social, physical, privacy, and
time-loss (Jacoby and Kaplan, 1972; Kaplan et al., 1974; Roselius, 1971) (see Table 1).
This study synthesizes the ECM, TAM, TPB and perceived risk to hypothesize a theoretical
continuance intention. Figure 2 illustrates the research model. It asserts that the continued
intention to use online shopping is a function of: satisfaction post-usage perceived usefulness
and perceived risk. The model further indicates satisfaction mediates the impact of
confirmation and post-usage perceived usefulness on the continued intention conduct online
7
shopping. The proposed constructs and hypotheses are supported by prior studies in
H2
Confirmation ECM
H5 Satisfaction H1
H3
Perceived
usefulness H4 Continued
Post-usage Intention
Perceived
usefulness
Pre-usage
Perceived H7 TAM
usefulness
H6 H14
H9 Initial
Attitude Intention
H10
H13
Perceived H8
ease of use H11
Perceived
risk
H12
Subjective
norms
TPB
Perceived
behavior
control
Prior marketing studies have found that the higher the users’ expectations, the higher are
their satisfaction (Oliver and DeSarbo, 1988). Moreover, the IT adoption literature has
consistently found that post-usage perceived usefulness is the most important determinant of
8
users’ adoption intentions (e.g., (Davis et al., 1989; Taylor and Todd, 1995b; Venkatesh,
2000)). As a result, the ECM posits users’ post-usage perceived usefulness of IT has a
positive effect on their intention to continue IT usage. Lastly, the ECM posits that the users’
experience, particularly when the users’ initial perceived usefulness is not concrete due to the
uncertainty over what to expect from using the IT (Bhattacherjee, 2001). Because online
H1. Users’ satisfaction with online shopping is positively related to their continuance
intention to shop online.
H3. Users’ post-usage perceived usefulness of online shopping is positively related to their
satisfaction toward online shopping.
H4. Users’ post-usage perceived usefulness of online shopping is positively related to their
continuance intention to shop online.
Consumers need to see online shopping as a useful information technology that can
improve their purchase efficiency, providing compelling advantages such as broader product
lines, faster transaction, and greater flexibility. Moreover, online shopping users need feel
that the online shopping is easy to use. Both pre-usage perceived usefulness and perceived
ease of use are beliefs that, according to TRA, will affect a user’s attitude and initial intention.
Thus, we have:
9
H6. Pre-usage perceived usefulness is positively related to behavioral attitude toward online
shopping.
H7. Pre-usage perceived usefulness is positively related to the initial intention toward online
shopping.
H8. Perceived ease of use is positively related to behavioral attitude toward online shopping.
In addition, TAM states that perceived usefulness will have a direct effect on user’s
behavioral intention, and perceived ease of use affects behavioral intention indirectly through
pre-usage perceived usefulness (Davis, 1989). That is, pre-usage perceived usefulness
mediates the effect of perceived ease of use on behavioral intention. Many empirical studies
have supported this argument (Venkatesh and Davis, 2000; Wu and Chen, 2005). Thus,
H9: Perceived ease of use is positively related to pre-usage perceived usefulness of online
shopping.
appraisal of the behavior in question’’ (Fishbein and Ajzen, 1975). According to the TPB,
attitude impacts users’ behavioral intention, which in turn influences their actual behavior.
When individuals form positive attitude towards online shopping, they will have a stronger
intention toward adopting it, and thus they are more likely to use it.
H10. Behavioral attitude toward online shopping is positively related to the initial intention to
shop online.
Subjective norm refers to ‘‘the perceived social pressure to perform or not to perform
the behavior’’ (Ajzen, 1991). In other words, subjective norm is related to the normative
beliefs about the expectation from other people. Many Internet users choose to shop online
because their friends shop online too, and they recommend it to them. Hence, we propose:
H11. Subjective norm is positively related to the initial intention toward online shopping.
10
Perceived behavioral control refers to ‘‘people’s perception of ease or difficulty in
performing the behavior of interest’’. It is associated with beliefs about the presence of
control factors that may facilitate or hinder the performance of the behavior in question
(Ajzen, 1991, 2002; Liao et al., 2006). In this case, although online shopping is a useful tool
to improve shopping efficiency, users still need to have the basic Internet skills to use it. Thus,
we posit that:
H12. Perceived behavioral control is positively related to the initial intention toward online
shopping.
The theory of perceived risk has been applied to explain consumer’s behavior in decision
making since the 1960s. Mitchell (1999) suggested that perceived risk is powerful at
explaining consumers’ behavior because consumers are more often motivated to avoid
the risk perceived by consumers could reduce their intention to buy through a particular
website. Similar logic should hold true for perceived risk toward continuance intention. Thus,
we propose that:
H13. Perceived risk is negatively related to the initial intention to use online shopping.
H14. Perceived risk is negatively related to the continuance intention to use online shopping.
4. Research methodology
We used a questionnaire survey with two parts to test our theoretical model. The first
part had questions measuring the constructs in the research model, while the second part had
demographic questions about the participants. Each item corresponding to the constructs was
measured using a seven-point Likert scale, with answer choices ranging from “disagree
11
strongly” (1) to “agree strongly” (7), and most of these items were adapted from the extant
literature. The scale items for perceived usefulness and perceived ease of use were adapted
from Davis(1989) and Bhattacherjee (2001), while the scale items for subjective norm,
perceived behavioral control, and attitude were adapted from the Taylor and Todd (1995a). In
addition, the items measuring perceived risk was adapted from Lee (2009), and the
After the questionnaire was drafted, it was first sent to three academic experts on online
shopping for their review, and then revised according to their comments and suggestions to
The questionnaire was pilot-tested by convenient sampling. There were 150 responses, of
which 123 were complete, giving a valid response rate of 82%, and the results of the pilot test
were evaluated by using Cronbach’s reliability and factor analysis. The reliability coefficient
was first calculated for the items of each construct, and the standard lower bound for
Cronbach’s alpha set at 0.7 (Hairs et al., 1998), with items that did not significantly
contribute to the reliability being eliminated. A factor analysis was then performed to
examine whether the items produced the anticipated number of factors and whether the
individual items were loaded on their appropriate factors. All items had high loadings on their
related factors and low cross-loadings on other factors, showing good convergent and
discriminate validities.
The participants had prior experience of using the Internet but had no experience of
online shopping. They aged from 21 to 60, and their educational backgrounds graduated from
junior school graduates to Ph.D. holders. Finally, 62% of the respondents were male. Table 1
12
These e-marketing courses last for three months. During this period, all the participants
received a two-stage survey. In the first stage survey, they were given a two-hour training
session by three research assistants to help understand the purchasing procedure of the online
shopping. Then, they were asked to just visit shopping stores but were not allowed to
purchase at that time. After this training, we conducted the questionnaire to measure subjects’
perceptions of pre-purchase beliefs for online shopping such as perceived risk, perceived
usefulness, attitude, subjective norm, perceived behavioral control, and initial intention. At
In the second stage survey, subjects who had participated in the first-stage survey were
measure subjects’ post-purchase beliefs for online shopping, such as feelings of confirmation,
invalid answers were eliminated, and a total of 302 valid responses were received. The results
showed that all the subjects accomplished the shopping procedure completely, namely
browsing, gathering product information, making product and price comparisons, making
purchase decision, and completing a transaction by offering payment and address information.
In addition, 288 of the participants shopped online more than once in the period studies.
5. Results
Anderson and Gerbing (1988). First, we examined the measurement model to measure
convergent and discriminant validity. We then examined the structural model to investigate
the strength and direction of the relationships among the theoretical constructs.
Convergent validity was assessed based on the criteria that the indicator’s estimated
13
coefficient was significant on its posited underlying construct factor. We evaluated the
measurement scales using the three criteria suggested by Fornell and Lacker (1981), namely:
(1) All indicator factor loadings ( ) should be significant and exceed 0.5
(3) Average variance extracted (AVE) by each construct should exceed the variance due to
measurement error for the construct (e.g., AVE should exceed 0.5)
The Cronbach’s alpha scores, shown in Table 2, indicated that each construct exhibited
strong internal reliability, which all the standard factor loading ( ) values in confirmatory
factor analysis of the measurement model exceeded 0.5 and were significant at p=0.001. In
addition, the composite reliabilities of constructs ranged from 0.81 to 0.93, and the AVE,
ranging from 0.57 to 0.83, was greater than the variance due to measurement error. Therefore,
Discriminant validity assesses the extent to which a concept and its indicators differs
from another concept and its indicators (Bagozzi et al., 1991). According to Fornell and
Larcker (1981), the correlations between items in any two constructs should be lower than the
square root of the average variance shared by items within a construct. As shown in Table 3,
the square root of the variance shared between the construct and its items was greater than the
correlations between the construct and any other construct in the model, satisfying Fornell
and Larckers’ (1981) criteria for discriminant validity. All diagonal values exceeded the
inter-construct correlations, and thus the results confirmed that our instrument had
14
Confirmation 0.31**
Satisfaction
0.33* 0.53**
0.22***
0.37***
Continued
Post-usage Intention
Perceived
usefulness
Perceived -0.11
Pre-usage risk
Perceived 0.21*
usefulness -0.41*
0.11* *
0.41*
Attitude Initial
0.19* Intention
Perceived 0.31*
ease of use
0.22*
0.25**
Subjective
norm
Perceived
behavior
control
We assessed the overall goodness of fit using the chi-square test. The chi-square test assesses
the adequacy of a hypothesized model in terms of its ability to reflect the variance and
covariance of the data. Due to its tendency to be sensitive to sample size, other fit indices
(namely GFI, AGFI, CFI, NFI, and RFI) were considered in conjunction with the chi-square.
For the statistical significance of parameter estimates, t values were used. The results of
structural equation modeling obtained for the proposed conceptual model revealed a ratio of
chi-square to the degree of freedom ( 2 / df ) of 2.04 ( p < 0.001), goodness-of-fit index (GFI)
15
of 0.91, adjusted goodness-of-fit index (AGFI) of 0.85, comparative fit index (CFI) of 0.95,
normed fit index (NFI) of 0.95, relative fit index (RFI) of 0.94, and root mean square error of
approximation (RMSEA) of 0.05. Generally, fit statistics greater than or equal to 0.9 for GFI,
NFI, RFI, and CFI indicate a good model fit (Hairs et al., 1998). Furthermore, RMSEA
values ranging from 0.05 to 0.08 are acceptable (Hairs et al., 1998), indicating that our model
fit was acceptable. The other fit indices, except AGFI, indicated that our proposed model
obtained an adequate model fit. The low AGFI values may have been due to the small sample
size used.
The fourteen hypotheses presented above were tested collectively using the structural
equation modeling (SEM) approach (Bagozzi et al., 1991), also performed using AMOS 6.
The path significance of each hypothesized association in the research model and variance
explained (R2 value) by each path were examined, and Fig. 3 and Table 4 show the
standardized path coefficients and path significances. All fourteen hypothesized associations
were strongly significant at p<0.05, except for one link between perceived risk and
continuance intention. The continuance intention to use online shopping in this study was
( 0.37 , p 0.001 ) and perceived rik ( 0.12 , p 0.12 ) and these variables together
In addition to its direct effects, post-usage perceived usefulness also had a significant indirect
effect on continuance intention ( 0.12 , p 0.05 ) via the satisfaction, explaining 18% of
the variance in the dependent variables. Satisfaction, in turn, was predicted by confirmation
Confirmation also had a small indirect effect ( 0.06 , p 0.05 ) on satisfaction, via the
16
post-usage perceived usefulness construct (see Table 4).
perceived behavioral control ( 0.25 , p 0.01 ), and together these variables explained
67% of the total variance. To further assess the significance of indirect effects of predictor
variables on intentions to use online shopping, a decomposition of the effects analysis was
conducted (see Table 5). A discussion of these effects for online shopping continuance is
6. Discussion
The results of this study provide support for the research model and for the hypotheses
regarding the directional linkage among the model’s variables. The overall explanatory power
of our research model had an R-square of 75% for initial intention to use online shopping and
an R-square of 80% for continued intention, suggesting that the extended ECM model is
shopping. Several significant results could be derived from our research framework, and
We examined the effects of satisfaction, perceived usefulness, and perceived risk on the
continuance intention with regard to online shopping. The results show that satisfaction is the
post-usage perceived usefulness ( 0.37 , p<0.05) as a significant but weaker predictor, and
17
the variance of continuance intention relative to the other predictors, which jointly explained
33% of the variance. Users who are dissatisfied with online shopping may stop using it,
despite having positive perceptions with regard to other elements of the experience. In other
Although both the post-usage and pre-usage perceived usefulness were identified in this
study as important determinants of initial intention and continuance intention toward online
shopping, the post-usage perceived usefulness has a stronger effect than pre-usage perceived
usefulness. This is consistent with Bhattacherjee (2001), which found there are often huge
changes then have an impact on subsequent cognitive processes. The pre-usage perceived
usefulness is a cognitive belief, formed potentially via second-hand information from referent
others, the media, advertising or other sources. These influence sources may be biased, and
hence user attitude potentially may be inaccurate, unrealistic, and uncertain. In contrast,
more realistic, unbiased, and less susceptible to change. In view of this, online store
practitioners should adopt a two-fold strategy in order to increase the initial adoption and
continued usage of online shopping: specifically, they should inform new (potential) users of
the potential benefits of online shopping use, and educate old (continued) users on how to use
online shopping more effectively so as to maximize their perceived usefulness with such
methods.
It is worth noting that the perceived risk has not significant influence on the continuance
intention. One possible explanation for this is that once users decide to continue to repurchase
online, they are satisfied with and /or perceive the usefulness of such purchases, and so are
18
willing to accept the related risks. In contrast, perceived risk has a significantly negative
impact on the initial intention to use online shopping. This implies that before adopting
online shopping, concerns about perceived risk are foremost in the minds of online
consumers, and thus may be the prime reason for not undertaking such purchases.
Consequently, practitioners need to develop for risk-reducing strategies that might assist in
The results of this study also support the idea that the user’s confirmation and post-usage
confirmation has the primary effect on satisfaction. Combined with the earlier finding that
satisfaction is the primary antecedent of continuance intention, these two results confirm the
While all the TAM and ECM variables are significant in this study, they implicitly assume
that behavior is volitional. However, online shopping users face several novel constraints,
such as the impersonal nature of the online environment, certain necessary resources and
skills (perceived behavioral control), and the influence of the opinions of other important
persons (subjective norms). These issues call for the inclusion of TPB in the online shopping
adoption model, and both subjective norms and perceived behavioral control are verified as
having a significant influence on continuance intention. This finding implies that when users
find that people around them have adopted online shopping services, they will be more
willing to use them. Online shopping providers may thus use a positive word-of-mouth
strategy to enhance the awareness of their services and promote their benefits. They may also
need to consider how to bring more positive experiences to their existing customers to retain
their acceptance, rather than relying on the mass media to deliver reinforcing message
(Bhattacherjee, 2000).
19
7. Implications
In terms of theory building, this study attempts to develop a new theory by grounding a
new variable in an integration of two schools of the nomological structure model of theory of
and applying them into a new context. It is important to note that the new variable, perceived
risk, is compatible with the TAM, TPB and ECM variables. This approach is likely to ensure
a stable theory development. Hence, the proposed model makes an important contribution to
The present study has two implications for future online shopping research. First, the
empirical results show that the unified model supports all the hypotheses and has good
explanatory power, implying that the integration of ECM, TAM, and TPB provides a model
with a theoretical basis to explain online shopping. This approach may provide an initial
blueprint for the further integration of other theoretical acceptance models. For example,
information technology (IT) research (Venkatesh et al., 2003) has already yielded many
competing models such as innovation diffusion theory (IDT), social cognitive theory (SCT),
expectation disconfirmation theory (EDT) and theory of reasoned action (TRA), each with
different sets of acceptance determinants. It is anticipated that this study may encourage other
Second, while the results show that the factors in TAM, TPB and ECM all have
satisfaction has the strongest effect. Since confirmation and expectations are critical
antecedents to satisfaction, future research may explore what factors influence these variables
and how they can be manipulated in order to improve eventual user experience with online
20
shopping, and hence its subsequent continuance.
The results of this study shed light on some important issues related to consumers’ initial
adoption and continuance intention toward online shopping. First, while the intention to use
perceived risk. This implies that controlling the risk of online shopping is particularly
important for managers as they decide how to allocate resources to retain and expand their
environment is a difficult thing. Therefore, online stores need to search for risk-reducing
strategies that might assist in inspiring high confidence in potential customers. This study
suggests that they should consider focusing on the prevention of fraud, identity theft, and
financial loss. For example, building secure firewalls to avoid intrusion, developing methods
for strengthening encryption, and authenticating websites are all measures that should be
undertaken. In addition, this study suggests that online shopping companies could develop
familiarity through advertising, and long-term customer service. It is worth noting that
because online shopping is a less verifiable and controllable environment, online shopping
customers usually have difficulty in asking for compensation when dissatisfaction for
products/services occurred. Thus, we suggest that online shopping companies should provide
customers with legal digital receipts or a guarantee for every transaction in order to increase
Several limitations and suggestions for future research are worth noting. First, some
threats to internal validity may need to be taken into account with a longitudinal study such as
21
this one, including history, maturation, testing, mortality and regression (Bhattacherjee and
proactive steps, as suggested by Bhattacherjee and Premkumar (2004). The time period
between pre-usage and usage stage is three months, and this period is short enough to avoid
the effects of history and maturation, but long enough to get rid of effect of testing. In
addition, our high response rates between both stages demonstrate a low mortality effect.
Second, because we collected the data for the independent and dependent variables from the
same respondents, concerns about common method bias could arise (Woszczynski and
Whitman, 2004), and we thus conducted Harmon’s one-factor test (Podsakoff et al., 2003) to
assess the risk of this. We entered all the variables into a factor analysis. These factors
emerged with the first factor, accounting for 23.4% of the variance in the variables. Because
more than a single factor emerged from the factor analysis and no general factor accounted
for the majority of the variance in those variables, we saw no evidence to suggest the
presence of common method variance bias. Third, about 62% of the respondents were male in
this empirical study, and thus the gender distribution was not symmetrical. Therefore, the
results of the current empirical study might tend to model the specific behavior of men, rather
than general behavior of all users. Much evidence has shown that gender differences can
cause discrepancies in the effects of satisfaction, perceived behavioral control, and subjective
norms on a user’s behavioral intention (Armitage et al., 2002; Liao et al., 2006). Other
references also show that the effects of ease of use and perceived usefulness on behavioral
intention can be moderated by gender difference (Gefen and Straub, 1997). Accordingly,
further research may be needed to examine the moderating effect of gender difference on the
22
References
Ajzen, I. (1991), "The Theory of Planned Behavior", Organizational Behavior and Human
Decision Processes, Vol. 50 No. 2, pp. 179-211.
Ajzen, I. (2002), "Perceived behavior control, self-efficacy, locus of control, and the theory
of planned behavior", Journal of Applied Social Psychology, Vol. 32 No. 1, pp. 1-20.
Ajzen, I. and Fishbein, M. (1980), Understanding attitudes and predicting social behavior.,
Englewood Cliffs, NJ: Prentice-Hall.
Anderson, E.W. and Sullivan, M.W. (1993), "The antecedents and consequences of
customer satisfaction for firms," Marketing Science, Vol. 12 No. 2, pp. 125-143.
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. 411-423.
Armitage, C.J., Norman, P. and Conner, M. (2002), "Can the theory of planned behavior
mediate the effects of age, gender and multidimensional health locus of control?",
British Journal of Health Psychology, Vol. 7 No. 2, pp. 299-316.
Bagozzi, R.P., Yi, Y. and Phillips, L.W. (1991), "Assessing Construct Validity in
Organizational Research", Administrative Science Quarterly, Vol. 36 No. 3, pp.
421-430.
Bearden, W.O. and Teel, J.E. (1983), "Selected determinants of consumer satisfaction and
complaint reports", Journal of Marketing Research, Vol. 20 No. 1, pp. 21-28.
Bhattacherjee, A. (2000), "Acceptance of Internet applications services: the case of electronic
brokerages.", IEEE Transactions on Systems, Man, and Cybernetics --- Part A:
Systems and Humans, Vol. 30 No. 4, pp. 411-420.
Bhattacherjee, A. (2001), "Understanding information systems continuance: An expectation
confirmation model", MIS Quarterly, Vol. 25 No. 3, pp. 351-370.
Bhattacherjee, A. and Premkumar, G. (2004), "Understanding changes in belief and attitude
toward information technology usage: a theoretical model and longitudinal test", MIS
Quaterly, Vol. 28 No. 2, pp. 229-254.
Cunningham, S.M. (Ed.). 1967. The major dimensions of perceived risk. In: Cox, D.F. (Ed.),
Risk Taking and Information Handling in Consumer Behaviour.: Harvard
University Press, Boston, MA.
Davis, F.D. (1989), "Perceived usefulness, perceived ease of use, and user acceptance of
information technology.", MIS Quarterly, Vol. 13 No. 3, pp. 319-340.
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989), "User acceptance of computer
technology: a comparison of two theoretical models", Management Science, Vol. 35
No. 8, pp. 982-1003.
Entrust (2008), "Pshing attack", http://www.entrust.com/phishing/index.htm, Vol. No., pp.
Featherman, M.S. and Pavlou, P.A. (2003), "Predicting e-services adoption: a perceived
risk facets perspective", International Journal of Human-Computer Studies, Vol. 59
No. 4, pp. 451-474.
Fishbein, M. and Ajzen, I. (1975), Belief, attitude, intention, and behavior: An introduction
to theory and research., MA: Addison-Wesley.
Fornell, C. and Larcker, D.F. (1981), "Evaluating Structural Equation Models with
Unobservable Variables and Measurement Error", JMR, Journal of Marketing
Research, Vol. 18 No. 1, pp. 39-47.
Forsythe, S.M. and Shi, B. (2003), "Consumer patronage and risk perceptions in Internet
shopping", Journal of Business Research, Vol. 56 No. 2, pp. 867-875.
Gefen, D., Karahanna, E. and Straub, D.W. (2003), "Trust and TAM in online shopping:
An integrated model", MIS Quaterly, Vol. 27 No. 1, pp. 51-90.
23
Gefen, D. and Straub, D.W. (1997), "Gender differences in the perception and use of email:
An extension to the technology acceptance model", MIS Quarterly, Vol. 21 No. 4, pp.
389-400.
Hairs, J.F., Anderson, R.E. and Tatham, R.L. (1998), Multivariate Data Analysis, 5th ed.
New York:Macmillian.
Hsu, M.H., Yen, C.H., Chiu, C.M. and Chang, C.M. (2006), "A longitudinal investigation
of continued online shopping behavior: An extension of the theory of planned
behavior. ", International Journal Human-Computer Studies, Vol. 64 No. 6, pp.
889-904.
Jacobs, P. (1997), "Privacy: what you need to know", InfoWorld, Vol. 19 No. 44, pp.
111-112.
Jacoby, J. and Kaplan, L., B (1972), the components of perceived risk, Association for
Consumer Research, Chicago.
Kaplan, L.B., Szybille, G.J. and et al. (1974), "Components of perceived risk in product
purchase: a cross validation", Journal of Applied Psychology, Vol. 59 No. 3, pp.
278-291.
Kuisma, T., Laukkanen, T. and Hiltunen, M. (2007), "Mapping the reasons for resistance to
Internet banking: A means-end approach", International Journal of Information
Management, Vol. 27 No. 2, pp. 75-85.
Lee, M.-C. (2009), "Factors influencing the adoption of internet banking: An integration of
TAM and TPB with perceived risk and perceived benefit", Electronic Commerce
Research and Applications, Vol. 8 No. 3, pp. 130-141.
Liao, C., Chen, J.-L. and Yen, D.C. (2006), "Theory of planning behavior (TPB) and
customer satisfaction in the continued use of e-service: An integrated model",
Computers in Human Behavior, Vol. 23 No. 6, pp. 2804-2822.
Liao, S. (1999), "The adoption of virtual banking: an empirical study", International Journal
of Information Management, Vol. 19 No. 1, pp. 63-74.
Lin, W.-B. (2008), "Investigation on the model of consumers' perceived risk --- integrated
viewpoint", Expert Systems with Applications, Vol. 34 No. 1, pp. 977-988.
Littler, D. and Melanthiou, D. (2006), "Consumer perceptions of risk and uncertainty and
the implications for behaviour towards innovative retail services: The case of Internet
Banking", Journal of Retailing and Consumer Services, Vol. 13 No. 6, pp. 431-443.
Lu, Y., Zhou, T. and Wang, B. (2009), "Exploring Chinese users' acceptance of instant
messaging using the theory of planned behavior, the technology acceptance model,
and the flow theory", Computers in Human Behavior, Vol. 25 No. 1, pp. 29-39.
Mitchell, V.W. (1999), "Consumer perceived risk:conceptualizations and models", European
Journal of Marketing, Vol. 33 No. 2, pp. 163-195.
Oliver, R.L. (1980), "A cognitive model for the antecedents and consequences of
satisfaction," Journal of Marketing Research, Vol. 17 No., pp. 460-469.
Oliver, R.L. (1993), "Cognitive, Affective, and Attribute Bases of the Satisfaction Response",
The Journal of Consumer Research, Vol. 20 No. 3, pp. 418-430.
Oliver, R.L. and DeSarbo, W.S. (1988), "Response Determinants in Satisfaction
Judgments", The Journal of Consumer Research,, Vol. 16 No. 4, pp. 495-507.
Peter, J.P. and Ryan, M.J. (1976), "An investigation of perceived risk at the brand level",
Journal of Marketing Research, Vol. 13 No., pp. 184-188.
Podsakoff, P.M., MacKenzie, S.B. and Lee, J.Y. (2003), "Common Method Biases in
Behavioral Research: A Critical Review of the Literature and Recommended
Remedies", Journal of Applied Psychology, Vol. 88 No. 5, pp. 879-903.
Reavley, N. (2005), "Securing online banking", Card Technology Today, Vol. 17 No. 10, pp.
12-13.
24
Roselius, T. (1971), "Consumer Rankings of Risk Reduction Methods", Journal of Marketing
Vol. 35 No. 1, pp. 56-61.
Soopramanien, D.G.R. and Robertson, A. (2007), "Adoption and usage of online shopping:
An empirical analysis of the characteristics of ''buyers'' ''browsers" adn "non-internet
shoppers"", Journal of Retailing and Consumer Services, Vol. 14 No. 1, pp. 73-82.
Szymanski, D.M. and Henard, D.H. (2001), "Customer satisfaction: a meta-analysis of the
empirical evidence", Journal of the Academy of Marketing Science, Vol. 29 No. 1, pp.
16-35.
Taylor, S. and Todd, P.A. (1995a), "Assessing IT usage: the role of prior experience", MIS
Quarterly, Vol. 19 No. 4, pp. 561-570.
Taylor, S. and Todd, P.A. (1995b), "Understanding information technology usage: a test of
competing models", Information Systems Research, Vol. 6 No. 2, pp. 144-176.
Venkatesh, V. (2000), "Determinants of perceived ease of use: integrating control, intrinsic
motivation, and emotion into the technology acceptance model.", Information Systems
Research, Vol. 11 No. 4, pp. 342-365.
Venkatesh, V. and Davis, F.D. (2000), "A theoretical extension of the technology
acceptance model: four longitudinal field studies", Management Science, Vol. 46 No.
2, pp. 186-204.
Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, M.D. (2003), "User acceptance of
information technology: Toward a unified view", MIS Quarterly, Vol. 27 No. 3, pp.
425-478.
Woszczynski, A.B. and Whitman, M.E. (Eds.). 2004. The problem of common method
variance in IS research: The Handbook of Information Systems Research. Idea
Group Inc., Hershey, PA, pp.66-77.
Wu, I.L. and Chen, J.L. (2005), "An extension of Trust and TAM model with TPB in the
initial adoption of online tax: An empirical study", International Journal of
Human-Computer Studies, Vol. 62 No. 6, pp. 784-808.
25
Appendix A. The questionnaire
26
Table 2: Construct reliability and convergent validity
27
Table 3: Correlation Matrices and Discriminant Validity
Construct Pre-usage Post-usage Perceived Subjective Perceived Attitude Initial Continued Confirm Satisfaction Perceived
perceived perceived ease of Norm Behavior intention Intention -ation risk
usefulness usefulness use control
Pre-usage 0.833
Perceived
usefulness
Post-usage 0.33 0.855
perceived
usefulness
Perceived ease 0.53 0.43 0.867
of use
Subjective norm 0.21 0.41 0.32 0.851
Perceived 0.24 0.34 0.34 0.35 0.856
behavioral
control
Attitude 0.26 0.63 0.51 0.54 0.39 0.85
Initial 0.31 0.22 0.42 0.51 0.31 0.42 0.84
intention
Continuance 0.27 0.69 0.55 0.67 0.51 0.62 0.53 0.90
intention
Confirmation 0.28 0.52 0.43 0.12 0.22 0.54 0.23 0.51 0.88
Satisfaction 0.43 0.61 0.23 0.20 0.20 0.58 0.33 0.24 0.23 0.77
Perceived risk -0.35 -0.23 -0.24 -0.23 -0.26 -0.62 -0.31 -0.24 -0.21 -0.02 0.85
Note: All correlations significant at p< 0.05 except where noted. Diagonal elements are square roots of average variance extracted
28
Table 4: Summary of hypotheses tests
29
Table 5. Direct, indirect and total effects --- Estimates
30