International Journal of Information Management 35 (2015) 4556

Contents lists available at ScienceDirect

International Journal of Information Management

journal homepage: www.elsevier.com/locate/ijinfomgt

Understanding the determinants of online repeat purchase intention

and moderating role of habit: The case of online group-buying in
Taiwan
Meng-Hsiang Hsu a , Chun-Ming Chang b, , Li-Wen Chuang c
a

Department of Information Management, National Kaohsiung First University of Science and Technology, Taiwan, ROC
Department of Tourism Information, Aletheia University, Taiwan, ROC
c
Department of Digital Media Arts & Design, Fuzhou University, Peoples Republic of China
b

a r t i c l e

i n f o

Article history:
Keywords:
Repeat purchase intention
Expectationconrmation model
Customer value
Habit
Trust
Satisfaction

a b s t r a c t
The success of online shopping depends on customers repeat purchasing. This study proposed a theoretical model by integrating the literature of expectationconrmation model and online shopping to test the
factors affecting repeat purchase intention in online group-buying. Data collected from 246 respondents
were used to test the proposed model. The results show that trust and satisfaction are the strong predictors of repeat purchase intention. The results also reveal that satisfaction has signicant inuence on trust,
whereas perceived value, conrmation, and website quality are the signicant antecedents of satisfaction. Our results report that website quality has positive inuence on perceived value. Finally, perceived
value exerts stronger effect on repeat purchase intention for high-habit customers, whereas trust and
satisfaction have higher inuence on repeat purchase intention for low-habit customers. Implications
and limitations are discussed.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Online group-buying is a business model of e-commerce that
enables customers to obtain volume discounts (Kauffman, Lai, & Ho,
2010). Many researchers have found that group-buying play a vital
role in e-commerce (Chen, Chen, & Song, 2007). A number of companies in the US and East Asia have adopted online group-buying
in their core business models (Kauffman & Wang, 2002; Kauffman,
Lai, & Ho, 2010). Despite the constant growth of interest in online
group-buying, customers loyalty in the online context is relatively
lower (Cheng & Huang, 2013). The retention of customers has
become an essential question for e-retailers to consider (Qureshi
et al., 2009; Shin, Chung, Oh, & Lee, 2013; Tsao, 2013). In this study,
Bhattacherjees (2001a) expectationconrmation model (ECM) is
applied to explore the determinants of repeat purchase intention
in online group-buying.
ECM is a well-researched model that has been used to predict behavior in various settings, including e-commerce (e.g.,
Bhattacherjee, 2001a, 2001b). Thus, it is reasonable to expect that
ECM could be used to explain online customer behavior. However,

Corresponding author. Tel.: +886 2 2621 2121x5210.

E-mail address: cmchang@mail.au.edu.tw (C.-M. Chang).
http://dx.doi.org/10.1016/j.ijinfomgt.2014.09.002
0268-4012/ 2014 Elsevier Ltd. All rights reserved.

previous literature argues that ECM should be extended by incorporating some theoretical perspectives to increase its predictive
power (Lin, Wu, & Tsai, 2005). By integrating the literature
of ECM (e.g., Hong, Thong, & Tam, 2006; Lin, Wu, Hsu, &
Chou, 2012) and online shopping (e.g., DeLone & McLean, 2004;
Zeithaml, Parasuraman, & Malhotra, 2002), we argue that perceived
usefulness in ECM is replaced by perceived value to reect customers perceptions about extrinsic and intrinsic benets in online
shopping and the viewpoint of website quality in DeLone and
McLean IS success model (D&M model) (DeLone & McLean, 2003)
can be used to reect customers perception about the quality of
the website (e.g., system quality, content quality, and service quality). Generally, integrating value and website quality into ECM
is in line with the assertions of Fornell, Johnson, Anderson, Cha,
and Bryant (1996) that customer expectations will affect perceived
value, perceived quality, and satisfaction. In addition, researchers
argue that satisfaction does not always predict customers continuous purchasing (Li, Browne, & Wetherbe, 2006) and trust is
a useful predictor of repeat purchase intention (Qureshi et al.,
2009). Accordingly, trust is be added to ECM to test its effect on
repeat purchase intention, following Casal, Flavin, and Guinalu
(20102011).
Researchers argue that the effects of the determinants of repeat
purchase intention may be contingent on the habit of online

46

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

shopping (Khalifa & Liu, 2007). Many studies have empirically

examined the moderating effect of habit on the link between trust
and repeat purchase intention (Chiu, Hsu, Lai, & Chang, 2012) and
the relationship between satisfaction and repeat purchase intention (e.g., Anderson & Srinivasan, 2003; Khalifa & Liu, 2007). Thus,
in this study, habit is included in ECM to test its moderating effects
on the linkages between repeat purchase intention and its determinants (i.e., perceived value, trust, satisfaction).
Overall, the objective of the study is to examine the antecedents
of online repeat purchase intention by integrating value, website
quality, trust, and habit into ECM. While past literature has examined the inuences of value, website quality, trust, and habit on
repeat purchase intention (e.g., Chiu et al., 2012; Gupta & Kim,
2007), to the best of our knowledge, little studies have been done
to combine these predictors of repeat purchase intention and to
test the moderating effects of habit on the relationships between
repurchase intention and its antecedents. In addition, few studies
have also been conducted to examine how transaction experience impacts these determinants, which in turn affects customers
repeat purchase intention. This study believes that the ndings of
this study may help both academics and practitioners gain insights
into how to promote customers repeat purchase intention.

2. Theoretical background
2.1. The online group-buying mechanism
Providing the lower prices to customers by accumulating a
large number of orders is the core concept of online group-buying
(Kauffman & Wang, 2002). Similar to online shopping, transactions
in the online group-buying include ve key processes, including
information searching (the process for gathering information about
products and prices), authentication (the process for verifying the
authenticity of the trading parties), price negotiation (the process for the discovery of a purchase and sale price for a product),
payment and settlement (the process for dening the terms and
method of payment), and post-transaction logistics (the process
for specifying and coordinating shipment of goods from the seller
and delivery to the customer) (Kauffman, Lai, & Lin, 2010). In particular, there are two types of pricing mechanisms in the process of
price negotiation, including posted price mechanism (sellers will
display the prices and they ask for the consumers decide whether
they would accept the prices or not) and dynamic pricing mechanism (customers can actively negotiate with the sellers to reach a
satisfactory price) (Kauffman & Wang, 2002). Moreover, the transactions in online-group buying could be initiated by an auction
initiator (a seller or customer) in some countries (Kauffman, Lai,
& Lin, 2010).
Recently, online group-buying business has become an interesting model for online transactions in some countries, such as US
(e.g., Goupon), Taiwan (e.g., iHergo and GOMAJI), and China (e.g.,
Taobao.com and Teambuy). Prior literature has also found that
online group-buying markets in East Asia are successful (Kauffman,
Lai, & Ho, 2010), since shopping in online group-buying in East Asia
is viewed as a leisure activity and the uncertainty about the nal
product price can be mitigated because prices are known when
customers decide to purchase products (Kauffman, Lai, & Ho, 2010).
While encouraging people to purchase online is an important
rst step toward success for online sellers, its eventual success still
depends on customers repurchasing (Bhattacherjee, 2001a; Shin
et al., 2013). However, prior literature argues that only a small
portion (about 1%) of online customers will return to make purchase (Gupta & Kim, 2007; Qureshi et al., 2009). It is therefore
important to explore the antecedents of repeat purchase intention in online group-buying (Qureshi et al., 2009). Recently, some

studies have found that structural embeddendness (Cheng &

Huang, 2013), relational embeddendness (Cheng & Huang, 2013),
satisfaction (satisfaction with website and satisfaction with sellers)
(Hsu, Chang, Chu, & Lee, 2014), and system quality (Cheng & Huang,
2013) are the key predictors of repeat purchase intention. However, understanding about the determinants of repeat purchase
intention in online group-buying is still limited. This study aims
to further explore the factors affecting repeat purchase intention
in online group-buying.
2.2. Expectationconrmation model
Building on expectationconrmation theory (Oliver, 1992),
Bhattacherjee (2001a) proposes ECM to explain a users intention
to continue using an IS (Limayem, Hirt, & Cheung, 2007). ECM holds
that users satisfaction and perceived usefulness are the predictors
of behavior intention, while users conrmation of expectation and
perceived usefulness will lead to users satisfaction. Conrmation of
expectation, on the other hand, also impacts perceived usefulness
(Bhattacherjee, 2001a). ECM has been widely used to examine the
factors affecting continued usage intention of an IS in the various
contexts, such as online banking, web portal, online services, and
mobile services (see Table 1). Since users IS continuance decision is
similar to consumers repurchase decision (Bhattacherjee, 2001a),
ECM is suited to be the theoretical base for this study to explore the
factors affecting customers post-purchase behavior.
Recently, researchers argue that some theoretical extensions
may be required for ECM in order to better explain users continuance behavior toward an IS in the various contexts (Lin et al.,
2005). Prior studies have begun to integrate some additional factors
into ECM to enhance the explanatory value of this model. Generally, their arguments can be categorized into three perspectives.
The rst perspective argues that perceived usefulness in ECM is
inadequate in predicting users continuance intention due to its
original denition. Prior study suggests that perceived usefulness
in ECM should be replaced with perceived value to reect to the
nature of cost/benet evaluation (Lin et al., 2012) and the effects
of extrinsic and intrinsic benet on behavioral intention simultaneously (e.g., Chea & Luo, 2008; Lin et al., 2005), since perceived
usefulness only reects the extrinsic benets resulted from IS usage
(Lin et al., 2012). The second perspective suggests that users IS
quality perception of should be included in ECM to examine the
link between technology features and continuance intention. In this
sense, perceived ease of use from the technology acceptance model
(TAM) has been added to ECM (e.g., Hong et al., 2006). Finally, the
third perspective treats relational capital as an important determinant of IS continuance. Empirical study has therefore integrated
trust, satisfaction, and commitment into ECM to explain customer
participation in online communities (Casal et al., 20102011).
Based on the statements mentioned above, we recognize that it
is appropriate to combine ECM and some additional factors to better understand the antecedents of behavioral intention in online
group-buying. In fact, the three perspectives proposed by ECM literature provide a sufcient theoretical base for us to identity the
factors that can be added into ECM. First, according to the perspectives of Lin et al. (2005) and Lin et al. (2012), perceived usefulness
in ECM is replaced by perceived value to test the role of extrinsic
and intrinsic benet of shopping experience in online shopping.
Second, although perceived ease of use is a strong predictor determining satisfaction and continuance intention in the ECM literature
(Hong et al., 2006), it only reects a users perception about the cognitive effort needed to learn and utilize an IS (Gefen, Karahanna, &
Straub, 2003). In order to capture the essential technical characteristics of an online shopping website, the aspect of website quality
proposed by D&M model is used in this study. Third, in order to
test the effect of relational capital and customer behavior, following

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

47

Table 1
Prior literature on expectationconrmation model.
Studies

Context

Dependent
variable(s)

Independent variable(s)

Key ndings

Bhattacherjee (2001a)

Online banking

Continuance
intention (CI)

Conrmation (CON), perceived

usefulness (PU), satisfaction (SAT)

Bhattacherjee (2001b)

Online brokerage

Continuance
intention (CI)

Casal et al. (2010-11)

Virtual
communities

Intention to follow
advice (IFD),
intention to use
community
product (ICP),

Conrmation (CON), perceived

usefulness (PU), satisfaction (SAT),
loyalty incentive (LI)
Conrmation (CON), perceived
usefulness (PU), satisfaction (SAT),
trust (TRT), affective commitment (AC),
participation (PA),

CON PU, PU SAT,

CON SAT, SAT CI,
PU CI
CON PU, CON SAT,
SAT CI, PU CI

Chea and Luo (2008)

Online service

Conrmation (CON), perceived

usefulness (PU), positive affect (PA),
negative affect (NA), satisfaction (SAT)

Hong et al. (2006)

Mobile Internet

Continuance
intention (CI),
recommendation
(REC), complain
(COM)
Continued IT usage
intention (CI)

Kang, Hong, and Lee (2009)

Online service

Continuance
intention (CI)

Conrmation (CON), perceived

usefulness (PU), perceived enjoyment
(PE), past use (PSU), regret (REG),
self-image congruity (SIC), satisfaction
(SAT)

Lee (2010)

E-learning

Continuance
intention (CI)

Limayem et al. (2007)

WWW

IS continuance
intention (CI)

Conrmation (CON), perceived

usefulness (PU), satisfaction (SAT),
perceived ease of use (PEU), attitude
(ATT), perceived enjoyment (PE),
concentration (CCT), subjective norm
(SN), perceived behavior control (PBC)
Conrmation (CON), perceived
usefulness (PU), satisfaction (SAT), IS
continuance usage (CU), habit (HAB),
frequency of past behavior (FPB),
comprehensive of usage (CHU)

Lin et al. (2005)

Web portal

IS continuance
intention (CI)

Lin et al. (2012)

Internet protocol
television

Continuance
intention of IPTV
(CI)

Recker (2010)

Process modeling
grammars

Continuance
intention (CI)

Conrmation (CON), perceived

usefulness (PU), satisfaction (SAT),
perceived ease of use (PEU),

Conrmation (CON), perceived

usefulness (PU), satisfaction (SAT),
perceived playfulness (PP)
Conrmation (CON), perceived benets
(PB), perceived sacrices (PS),
perceived value (PV), satisfaction (SAT)
Conrmation (CON), perceived
usefulness (PU), satisfaction (SAT),
perceived ease of use (PEU)

CON PU, PU SAT,

CON SAT, SAT TRT,
TRT AC, SAT PA,
SAT AC, PU PAT,
TRT IFD, AC PAT,
PAT ICP
CON PU, CON SAT,
CON PA, CON ()NA,
PU SAT, SAT ()COM,
SAT CI
CON PU, CON SAT,
CON PEU, PEU SAT,
PU CI, PEU CI,
PEU PU, SAT CI
CON PU, SIC PU,
CON PE, SIC PE,
PSU PE, PU SAT,
PE SAT, PSU CI,
PE CI, SIC CI, PU CI,
REG ()CI, SAT CI
CON PU, CON SAT,
PUSAT, PU CI,
PEU PU, PEU ATT,
PE ATT, ATT CI.
SAT CI, CCT CI,
SN CI, PBC CI
CON PU, CON SAT,
PU SAT, PU CI,
CI CHU, SAT CI,
SAT HAB, FPB HAB,
CHU HAB,
HAB*CI ()CHU
CON PU, CON SAT,
CON PP, PU CI,
PP SAT, PP CI, SAT CI
CON PB, CON SAT,
PU SAT, PS ()PV,
PV CI, PB SAT,
SAT CI
CON PU, CON SAT,
PU SAT, PU CI,
PEU SAT, PEU PU,
PEU CI, PU CI.
SAT CI

positive inuence; () negative inuence.

Casal et al. (20102011), we argue that ECM could be extended by

incorporating trust to test the inuence of relationship quality in
online shopping. In addition, previous literature argues that habit
exerts a moderating effect on the link between repeat purchase
intention and its antecedents (e.g., trust and satisfaction) (Anderson
& Srinivasan, 2003; Chiu et al., 2012). In this study, we treat habit as
a moderating variable that may impact the inuences of perceived
value, satisfaction, and trust on repeat purchase intention.
2.3. Prior studies on repeat purchase intention
Repeat customers are the main source of prot for retail business (Gupta & Kim, 2007). In this sense, how to entice customers
repeat purchase has become an important concern for any seller
in the retailing context (Qureshi et al., 2009). This is an even more
challenging issue for online sellers, since the spatial and temporal separation in the online environment prevents customers from

assessing the trustworthiness of the sellers and checking product

quality carefully and will make customers reluctant to engage in
online transactions (Pavlou, Liang, & Xue, 2007). Researchers have
found that the e-commerce market is still small despite the constant growth in the past decade (Qureshi et al., 2009). In order
to facilitate online repeat purchase intention, several studies have
identied the determinants of online repeat purchase intention
(see Table 2).
As shown in Table 2, the studies on repeat purchase intention
were presumably conducted in the context of online stores, such as
bookstores and general online shopping websites. The studies in the
table also demonstrate that some factors such as website quality,
satisfaction, trust, and value are the key predictors of repeat purchase intention and habit will exert moderating effect on the link
between repeat purchase intention and its antecedents (e.g., trust).
However, little studies have combined these factors to test their
effects on repeat purchase intention simultaneously. In addition,

48

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

Table 2
Prior literature on repeat purchase intention.
Studies

Context

Dependent
variable(s)

Independent variable(s)

Key ndings

Chiu et al. (2012)

Online auction

Repeat purchase
intention (RPI)

Trust (TRT), value (VAL), familiarity

(FAM), satisfaction (SAT), habit
(HAB)

Gupta and Kim (2007)

Online book store

Repurchase
intention (RI)

Convenience (CON), pleasure (PLE),

perceived price (PP), perceived
value, transaction experience (TE)

Khalifa & Liu (2007)

Online shopping

Online repurchase
intention (RI)

Kim and Gupta (2009)

Online book store

Qureshi et al. (2009)

Online shopping

Purchase intention
(PI)
Repurchasing
intention (RI)

Perceived usefulness (PU), online

shopping satisfaction (SAT), online
repurchase habit (HAB), online
repurchase experience (EXP),
Perceived risk (PR), perceived price
(PP), perceived value (PV)
Perceived Website Quality (PWQ),
perceived capability of order
fulllment (PCOF), reputation
(REP), trust in vendor (TRT)

VAL HAB, FAM HAB,

SAT HAB, VAL TRT,
VAL SAT, TRT RPI,
HAB RPI,
TRT*HAB ()RPI
PV RI, PP ()RI,
CON RI, PLE RI,
PP ()PV, CON PV,
PLE PV,
PU SAT, PU RI,
SAT RI, HAB SAT,
EXP SAT
PP ()PI, PV PI,
PR ()PV, PP ()
PWQ TRT, PCOF TRT,
REP TRT, PWQ RI,
TRT RI

positive inuence; () negative inuence.

most studies in the table lack the proper theoretical foundation to

explain how the transaction experience impacts customers repeat
purchase intention by modifying the determinants of repeat purchase intention. Finally, although the moderating role of habit has
been examined by prior studies (e.g., Chiu et al., 2012; Khalifa & Liu,
2007), little studies have been done to test the moderating effect
of habit on the link between perceived value and repeat purchase
intention. Little attention has been paid to compare the moderating effects of habit on the inuences of value, satisfaction, and trust
on repeat purchase intention as well. Thus, by integrating value,
website quality, trust, and habit into ECM, the critical gaps in the IS
literature could be lled.
2.4. The perspective of perceived value
Perceived value refers to ones overall assessment of a service
based on the ratio of perceived benets and perceived costs (Lin &
Wang, 2006). In the study, perceived value is dened as benets
received from online transaction given the time and effort spent. In
general, perceived value can be classied into two distinct dimensions, including utilitarian and hedonic value (Chiu et al., 2012).
Utilitarian value refers to the acquisition of products in an efcient manner, whereas hedonic value reects the value received
from the multisensory, fantasy and emotive aspects of shopping
experience (Chiu et al., 2012). In this study, we treat perceived
value as a construct composed of utilitarian value and hedonic
value. Furthermore, Perceived value is found to be a strong predictor of satisfaction (e.g., Chiu et al., 2012; Lin & Wang, 2006).
One basic assumption is that perceived value has positive impact
on satisfaction. In addition, the viewpoint of ECM asserts that conrmation of expectation will impact perceived benets. Similarly,
Fornell et al. (1996) asserts that customer expectations will affect
perceived value. Hence, it is reasonable to believe that conrmation
of expectations will impact perceived value as well.
2.5. The role of website quality
Online transaction is generally carried out on a web site and thus
website quality plays an important role in determining the success
of online shopping. From the perspective of D&M model, website
quality focuses on the user perception of website design quality
and it can be divided into three distinct dimensions: system quality,
information quality, and service quality (DeLone & McLean, 2003).

System quality refers to technical quality of information system,

while information quality is the quality of information (DeLone &
McLean, 2003). Service quality, on the other hand, refers to overall
support delivered by service provider (DeLone & McLean, 2003).
In this study, website quality is treated as a construct composed
of system quality, information quality, and service quality. D&M
model asserts that website quality is a vital determinant of satisfaction (DeLone & McLean, 2003). In addition, Zeithaml et al. (2002)
suggest that when users obtain actual usage experiences of a website, the perception of website quality will be formed, which in
turn impacts their value perception. Fornell et al. (1996) also suggest that customer expectations will have positive inuence on
perceived quality. Based on the above arguments, this study proposes that website quality mediates the inuence of conrmation
on value.
2.6. The role of trust
Trust refers to a condence/belief a person has that others in an
exchange will behavior as he/she expects (Gefen et al., 2003) and
would not act opportunistically by taking advantage of the situation (Qureshi et al., 2009). Similar to Pavlou et al. (2007), trust in
this study is dened as a customers intention to accept vulnerability based on his/her beliefs that transaction with a seller will meet
his/her expectations. Researchers suggest that trust is a mechanism that can mitigate the uncertainty of online transaction and
thus facilitates customers intention to engage in online exchange
relationship (Pavlou et al., 2007; Qureshi et al., 2009). Many studies also argue that satisfaction is a strong predictor of trust (Casal
et al., 20102011; Shin et al., 2013). Thus, trust is proposed to be
the antecedent of repeat purchase intention and satisfaction will
promote the development of customers trust in sellers.
2.7. The moderating role of habit
Habit is commonly dened as learned sequences of acts that
have become automatic responses to specic situations, which may
be functional in obtaining certain goals or end states (Verplanken
et al., 1997). Researchers also postulate that habit is a behavioral
tendency resulting from past experience (Khalifa & Liu, 2007) and
people will not devote rational evaluation prior to engaging in
social or economic behavior (Khalifa & Liu, 2007). In the context
of online shopping, habit is dened as an automatic behavioral

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

Value
H1

Habit
H11

H3
H5
H4

Confirmation

H6
H7

H2

Website
Quality

H8

H12

Satisfaction

H9

H13

Repeat
Purchase
Intention
H10

Trust

Fig. 1. Research model.

reaction that is stimulated by a condition/environment cause without a thinking or conscious mental process due to the cumulate past
experience connection between the shopping behavior and satisfactory results (Chiu et al., 2012; Limayem et al., 2007; Wood &
Neal, 2007; Wood, Quinn, & Kashy, 2002).
According to previous literature (Chiu et al., 2012; Khalifa & Liu,
2007), the link between habit and repeat purchase intention can be
categorized as two perspectives. The rst perspectives asserts that
habit exerts a direct effect on repeat purchase intention, while the
second perspective asserts that habit moderates the relationship
between repeat purchase intention and the antecedents of repeat
purchase intention. The focus of this study is the moderating role of
habit, since the direct effect of habit on repeat purchase intention
has been tested by previous studies (e.g., Chiu et al., 2012; Khalifa
& Liu, 2007). Thus, examining the changes in the effects of determinants of repeat purchase intention under the different strength of
habit may further account for the role of habit (Khalifa & Liu, 2007).
Past literature has found that habit will moderate the inuences
of trust and satisfaction on repeat purchase intention. For example,
Chiu et al. (2012) nd that importance of trust will decrease as habit
increases over time. Khalifa and Liu (2007) assert that satisfaction
may not necessarily lead to intention to return to an Internet store
when a habit of online shopping has not been formed. Anderson
and Srinivasan (2003), on the other hand, posit that habit will lower
the link between satisfaction and customer loyalty. Accordingly, we
recognize that the effects of the determinants of online repeat purchase intention may be contingent on the habit of online shopping
(Khalifa & Liu, 2007). In this study, we aim to compare the moderating inuence of habit on the linkages between repeat purchase
intention and its determinants (i.e., value, satisfaction, and trust).

49

be employed to examine the roles of technical characteristics and

human relationships in online group-buying. The rest of this section
will dene the constructs and develop the hypotheses presented in
the model.
Conrmation refers to a cognitive concept representing the
extent to which consumers ex ante expectations of service/product
use were met in reality and an evaluation process (Lee & Kwon,
2011) In ECM, conrmation implies the realization of expected benets, while disconrmation denotes that perceived benets fail to
achieve expectations (Bhattacherjee, 2001a). Moreover, from the
perspective of cognitive dissonance theory (Festinger, 1957), people tend to adjust their benet perception according to reality. If
they nd that their initial expectations of benets are not consistent with their benet perception (Bhattacherjee, 2001a; Chea &
Luo, 2008). The empirical evidence supporting the inuence of conrmation on perceived benets has been provided by prior studies
(e.g., Bhattacherjee, 2001a; Chea & Luo, 2008; Lee, 2010; Recker,
2010). Thus,
H1. Conrmation of expectation is positively related to customer
value.
The standpoint of SERQUAL asserts that website quality can be
considered as the difference between expected website quality and
perceived website quality (Devaraj, Fan, & Kohli, 2002; Zeithaml
et al., 2002). That is, website quality perception is the result of
the evaluation of website quality and expectation (Devaraj et al.,
2002). The perspective of cognitive dissonance theory (Festinger,
1957) also suggests that customers will adjust their website perception according to reality. By synthesizing the above standpoints, we
propose that conrmation of expectations will affect a customers
perception about website quality.
H2. Conrmation of expectations is positively related to website
quality.
Prior literature asserts that website quality is a key predictor
of perceived value (Tsao, 2013; Wu & Wang, 2006). From the TAM
(Davis, 1989), when a customer believes that getting product information and purchasing products/services from a website is free of
effort, perceived value will be increased. Website quality has been
shown to inuence perceived value in the previous literature (e.g.,
Wu & Wang, 2006). We therefore propose the following hypothesis.
H3. Website quality is positively related to perceived value.

3. Research model and hypotheses

Based on the theoretical background, the research model and its
hypotheses are shown in Fig. 1. In general, integrating perceived
value, perceived website quality, and trust into ECM are useful
to explore the determinants of online group-buying. First, online
group-buying has been seen as a type of collaborative online
shopping in which a customer shops with shopping partners in
the virtual context. Customers generally tend to use collaborative
online shopping websites to receive hedonic benets by purchasing products with shopping partners (Kim, Suh, & Lee, 2013; Liu,
Brock, Shi, Chu, & Tseng, 2013). Furthermore, online group-buying
is considered as an efcient way to purchase products as well (Liu
et al., 2013). Thus, perceived value could be used to reect customers perceptions about utilitarian and hedonic value of online
group-buying. Second, shopping in online group-buying involves
the interactions between online sellers and customers. Prior literature suggests that website quality and relationship quality are
important for remaining repeat purchase intention (Gefen et al.,
2003). Hence, website quality, trust and satisfaction are suited to

Satisfaction reects customers pleasure or disappoint resulting from comparing perceived performance with their expectations
(Chiu et al., 2012). Thus, when customers fell that perceived benets outperform expectations, satisfaction will result (Recker, 2010).
In the online context, past studies have found that conrmation
affects consumer satisfaction signicantly (Lee & Kwon, 2011;
Limayem et al., 2007; Lin et al., 2005). Therefore,
H4. Conrmation is positively related to satisfaction.
ECM also holds that customer satisfaction is the result of a customers perception of expected benets. From the perspective of
social science, affective response is triggered by cognitive though
process (Chiu et al., 2012). Generally, perceived value is considered as cognition-based construct capturing the benet-sacrice
discrepancy, whereas satisfaction is an affective evaluation process (Chiu et al., 2012; Lin & Wang, 2006). Thus, perceived value is
treated as the determinant of satisfaction. Previous literature has
provided empirical evidence to support the relationship between

50

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

perceived value and satisfaction (Chiu et al., 2012; Lin & Wang,
2006; Tsao, 2013). Thus,

positive inuence of trust on repeat purchase intention has been

tested empirically by many studies (e.g., Chiu et al., 2012; Shin et al.,
2013). Thus,

H5. Perceived value is positively related to satisfaction.

H10. Trust is positively related to repeat purchase intention.
Prior literature also nds that website quality is an important
determinant of customer satisfaction (Kim, Xu, & Koh, 2004; Teo,
Srivastava, & Jiang, 20082009). In general, satisfaction reects a
sellers ability to meet a customers expectations (Kim et al., 2004).
In the online shopping context, when a customer uses a website
to gather information or purchase products/services successfully,
he/she may consider that this website provides ease and prompt
access to quality information, and is able to be used to complete
transaction. The experience of transaction will thus increase customer satisfaction (Kim et al., 2004). Prior studies have provided
empirical evidence to support the inuence of website quality on
satisfaction (Kim et al., 2004; Shin et al., 2013; Udo, Bagchi, & Kirs,
2010). Thus,
H6. Website quality is positively related to satisfaction.
In the e-commerce setting, satisfaction refers to a customers
emotional reaction to the transactions experience with online sellers (Kim et al., 2004), while trust reects a customers expectations
that online sellers will behave in a dependable and acceptable
manner (Pavlou & Fygenson, 2006). Thus, when a customer feels
satised with an online transaction, he/she may consider that an
online seller is able to meet his/her expectations (Kim et al., 2004).
From the viewpoint of trust-building process, the ability to fulll a
customers expectations is a key driver for the development of trust
(Doney & Cannon, 1997). Thus, we may propose that satisfaction
will have positive effect on trust.
H7. Satisfaction is positively related to trust.
According to social exchange theory (Blau, 1964), people will
perform a specic behavior based on an expectation that they will
received some kind of reward. Similarly, prior literature considers that customers will conduct transaction with sellers whose
products/services can offer maximum value (Gupta & Kim, 2007).
Previous literature has provided empirical evidence to support the
link between perceived value and repeat purchase intention (e.g.,
Gupta & Kim, 2007), Hence,

As mentioned above, social exchange theory asserts that an individuals behavior is driven by expected benets. In this regard,
uncertainty, the possible reverse outcome, is regarded as a key
factor that may inhibit customers intention to engage in online
transactions (Pavlou et al., 2007). Prior literature argues that the
formation of habit requires a stable context (Limayem et al., 2007).
This implies that a customers behavior could be performed automatically when uncertainty perception is mitigated. Therefore, we
may expect that a strong habit will increase the effect of perceived
value on repeat purchase intention. Thus,
H11. Habit will increase the inuence of perceived value on repeat
purchase intention.
Habit is considered as a factor that will moderate the inuence
of satisfaction on repeat purchase intention positively. As Khalifa
and Liu (2007) suggest, if customers have same level of satisfaction
with an online store, a customer who has high levels of habit will
be more likely to repurchase from the same online store than those
without such habit. Khalifa and Liu (2007) also provide an empirical
evidence to support that habit positively moderate the inuence of
satisfaction on repeat purchase intention. Therefore,
H12. Habit will increase the inuence of satisfaction on repeat
purchase intention.
Habit has been also considered as a major factor moderating the
inuence of trust on repeat purchase intention (Chiu et al., 2012).
In general, habit is an automatic behavioral response resulting
from frequent performance in a stable context (Khalifa & Liu, 2007;
Ouellete & Wood, 1998). When online shopping behavior is repeatedly executed in a stable context and becomes habitual, the need
to engage in the cognitive evaluation of the online sellers trustworthiness will be suppressed (Chiu et al., 2012). Past literature
has found that habit exerts a moderating role in the relationship
between trust and repeat purchase intention (Chiu et al., 2012).
Therefore, we propose the following hypothesis.

H8. Perceived value is related to repeat purchase intention.

Previous studies postulate that customers repeat purchase
intention is determined by the level of satisfaction (e.g., Liu, Guo,
& Lee, 2011; Shin et al., 2013). Repeat purchase intention refers
to the subjective probability that a consumer will continue to purchase products/services from online sellers in the future (Chiu et al.,
2012). Generally speaking, satised customers usually possess
strong repeat purchase intention, while dissatised customers are
more likely to switch to another competitor (Lin & Wang, 2006).
Past studies also provide empirical evidence to support the link
between satisfaction and repeat purchase intention (e.g., Shin et al.,
2013; Tsao, 2013). Therefore, we propose the rst hypothesis.
H9. Satisfaction is positively related to repeat purchase intention.
It is well established that trust plays a vital role in the online
transaction process, because it can provide customers with high
expectations that sellers will be able to accomplish their tasks on
the website (Gefen et al., 2003). The lack of trustworthiness and
condence in sellers will reduce the motivation to continue to
engage in online transactions (Li et al., 2006; Shin et al., 2013). The

H13. Habit will reduce the inuence of trust on repeat purchase

intention.
4. Research methodology
4.1. Survey administration
In order to test our proposed model, we collected data from
the consumers of Groupon (www.GROUPON.com.tw) in Taiwan.
Groupon in Taiwan was chosen because online group-buying has
become a popular transaction mechanism in Taiwan and it is one
of the well-known online group-buying websites. The recent surveys show that more than 22.3% of Internet users in Taiwan have
purchase experience in the group-buying websites in Taiwan in
2010 (iThome, 2011) and the market share of Groupon in the online
group-buying market of Taiwan has reached 40 percent as of May
2012 (Business-Next, 2012). This website offers four categories of
commodity, such as sports, hotel, and leisure activities, food and
drink, services (e.g., auto detailing) and training courses, and beauty
and SPAs. In general, all the group-buying auctions will be initiated
by Groupon and the prices of these commodities will be posted on

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

51

Table 3
Demographic information about the respondents (N = 246).
Measure

Items

Frequency

Percentage (%)

Signicant value for

non-response bias test

Gender

Male
Female

61
185

24.8
75.2

0.118

Age

Below 24
2535
3645
Above 45

104
123
14
5

42.3
50
5.7
2

0.068

Education

High school
College (2 or 4-year)
Graduate or above

4
180
62

1.6
73.2
25.2

0.104

Shopping frequency (per month)

12
34
56
Above 6

88
73
62
23

35.8
29.7
25.2
9.3

0.235

the website that allows customers to decide whether they want to

accept these prices within a xed time period.
In order to target respondents, a banner with a hyperlink connecting to the Web survey was posted on a number of bulletin board
systems (BBS) and Facebook of Groupon between August 2011
and September 2011. Members with online group-buying experience were invited to support this survey. The respondents were
instructed to answer all of the questions based on their shopping
experience with the Groupon. The rst page of Web questionnaire
stated the purpose of this study and assured the condentiality. At the end of data collection, a total of 246 questionnaires
were collected for further data analysis. Table 3 lists the demographics of the respondents. The demographics of respondents are
similar to a recent survey conducted by Market Intelligence and
Consulting Institute (2011) that most of the online group-buying
buyers in Taiwan were females and the years of age ranged from
20 to 29, indicating that the sample in the study represents overall
online group-buying auction users in Taiwan. In addition, the nonresponse bias test was performed by comparing the early and late
respondents on demographic variables such as gender, age, education, and shopping frequency using t-test. Table 3 show that none of
the tests were statistically signicant, indicating that non-response
bias was not a main concern for this study (Armstrong & Overton,
1976).
4.2. Measurement development
The questionnaire was developed by adapting measures had
been validated by prior literature. Repeat purchase intention was
measured with items adapted from Khalifa and Liu (2007) and
Chiu et al. (2012), while the items used to measure satisfaction
were adapted from Chiu et al. (2012). Trust was assessed using
items adapted from Pennington, Wilcox, and Grover (20032004).
Website quality was measured using items adapted from Wu and
Wang (2006) and Teo et al. (20082009). Items for measuring
perceived value were adapted from Chiu et al. (2012). Conrmation was assessed using items adapted from Lee (2010), Limayem
et al. (2007) and Bhattacherjee (2001a). The items used to measure habit were adapted from Chiu et al. (2012) and Limayem et al.
(2007). For all measures, a ve-point Likert scale was used with
anchors ranging from strongly disagree (1) to strongly agree (5).
The questionnaire items are listed in Table 3.
5. Data analysis and results
In this research, partial least square (PLS) was used for data
analysis, since it places minimum restrictions on samples size,

measurement scales, and residual distributions (Pavlou &

Fygenson, 2006). We used a two-step approach to conduct
data analysis. First, the reliability and construct validity was
assessed by conducting conformation factor analysis (CFA) in the
rst step. Then, the structure model was examined in the second
step. In the study, we used SmartPLS Version 2.0 M3 (Ringle,
Wende, & Will, 2005) in our data analysis.
5.1. Measurement model
The adequacy of the measurement model was evaluated based
on the criteria of reliability, convergent validity, and discriminant
validity. First, reliability was examined based on the composite reliability values. Table 4 shows that all of the values are above 0.7
(Gefen, Straub, & Boudreau, 2000), indicating adequate composite reliability. Second, Table 4 show that all of the items exhibit a
loading higher than 0.7 on their respective constructs, and Table 4
shows that all of the AVEs range from 0.59 to 0.84. As a result,
results indicate the adequate convergent validity (Fornell & Larcker,
1981). Third, Table 5 lists the correlations among constructs, with
the square root of the AVE on the diagonal. The results demonstrate
the accepted discriminant validity (Fornell & Larcker, 1981). Finally,
a second way to evaluate convergent and discriminant validity
is to examine the factor loadings of each construct. As shown in
Appendix A, the results show that the loading of each indicator
on its assigned construct is larger than its loading on any other
construct, conrming the adequate convergent and discriminant
validity.
Moreover, in order to assess the concerns of multicollinearity,
we calculated the VIF values for all of the constructs in the mode.
According to Hair, Black, Babin, Anderson, and Tatham (2006), multicollinearity results when VIF exceeds 10. Our results show the
VIF did not exceed 2.01, thus, the problem of multicollinearity
does not inuence the results. Furthermore, since our data were
collected through self-report measures, we used Harmans onefactor test, for which all of the principal constructs were entered
into a principal component analysis to test common method bias
(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). The results show
that the rst (largest) factor accounts for 40.69% of it. Thus, no
single factor accounts for the majority of the covariance in the
data. In addition, following Liang, Saraf, Hu, and Xue (2007), we
included a common method factor whose indicators included all
the principal constructs indicators in the PLS model. Then we calculated each indicators variances substantively explained by the
principal construct and by the method (Liang et al., 2007). The
results demonstrate that the average indicators substantive variance is 0.70, while the average method based variance is 0.008. In

52

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

Table 4
Measurement Items.
Construct

Item

Factor loading

Conrmation (CON) (composite reliability = 0.87)

My experience with using the Groupon was better than what I expected.
CON1
CON2
The benet provided by the Groupon site was better than what I expected.
CON3
Overall, most of my expectations from using the Groupon were conrmed.

0.81
0.86
0.83

Perceived value (PV) (composite reliability = 0.85)

Using the Groupon is a convenient way to shop.
PV1
I can use Groupon site to shop anywhere and anytime.
PV2
I nd shopping at the Groupon stimulating.
PV3
I think that purchasing products from the Groupon is interesting.
PV4

0.75
0.74
0.80
0.79

Website quality (WQ) (composite reliability = 0.91)

The Groupon is easy to use.
WQ1
This Groupon is well designed for users.
WQ2
The Groupon provides sufcient information.
WQ3
This Groupon provides reliable information.
WQ4
WQ5
The Groupon provides up-to-date information.
WQ6
The Groupon provides dependable services.
WQ7
The Groupon gives prompt service to consumers.

0.74
0.78
0.79
0.79
0.79
0.78
0.76

Satisfaction (SAT) (SAT) (composite reliability = 0.90)

I feel good regarding my decision to purchase products from Groupon.
SAT1
SAT2
I think purchasing products from Groupon is a good idea.
I am satised with the experience of purchasing products from Groupon.
SAT3

0.90
0.82
0.86

Trust (TRT) (composite reliability = 0.93)

The Groupon is trustworthy
TRT1
The Groupon is known as one who keeps promises and commitments.
TRT2
The Groupon always keep members best interests in mind.
TRT3

0.89
0.93
0.89

Habit (HAB) (composite reliability = 0.88)

HAB1
Shopping at the Groupon is something I do frequently.
Shopping at the Groupon is nature to me.
HAB2
Shopping at the Groupon is something I do without thinking.
HAB3

0.85
0.87
0.84

Repeat purchase intention (RPI) (composite reliability = 0.94)

RPI1
If I could, I would like to continue using GROUPON site to purchase products
I plan to continue using GROUPON site to purchase products in the future.
RPI2
RPI3
It is likely that I will continue purchasing products from GROUPON site in the future.

0.94
0.94
0.86

addition, all the method factor loadings are not signicant. Thus,
we may contend that common method bias is not a major concern
for this study.
5.2. Structural model
The theoretical model and hypothesized relationships were estimated using the bootstrap approach with a sample size of 500 to
generate t-values and standard errors for determining the signicance of paths in the structural model. In order to test moderating
effect of habit, we use the subgroup analysis method to test the
model in full and each subgroup, following Ahuja and Thatcher
(2005) and Chang, Hsu, Hsu, and Cheng (2014). In this study, the
groups were divided into high-habit subgroup (N = 105) and lowhabit subgroup (N = 141) using the median (Baron & Kenny, 1986;

Chiu et al., 2012). Fig. 2ac summary the results of structural model
test.
As shown in Fig. 2a, conrmation inuences perceived value
and website quality signicantly, thereby validating H1 and
H2 ( = 0.288, 0.556; t = 4.612; 8.120, respectively). In addition,
coefcient between conrmation and satisfaction is signicant
( = 0.273, t = 4.787), thus supporting H3. Website quality exerts
positive effect on perceived value, ( = 0.484, t = 7.256), thus H4
is supported. Perceived value and website quality have signicant impacts on satisfaction as well ( = 0.314, 0.159; t = 4.673;
2.273, respectively). The results support H5 and H6. Furthermore, satisfaction exerts a positive effect on trust, while perceived
value has signicant inuence on repurchase intention ( = 0.467,
0.421; t = 5.783; 6.037), indicating H7 and H8 is supported. On the
other hand, satisfaction and trust have positive impacts on repeat

Table 5
Correlations of latent variables and AVE.
Construct

Mean

STD

AVE

CON

PV

WQ

SAT

TRT

HAB

RPT

CON
PV
WQ
SAT
TRT
HAB
RPT

4.16
4.34
4.43
4.29
4.33
4.36
4.37

0.58
0.51
0.55
0.56
0.63
0.54
0.56

0.69
0.59
0.60
0.74
0.82
0.72
0.84

0.83
0.56
0.56
0.54
0.50
0.54
0.57

0.77
0.65
0.57
0.41
0.50
0.59

0.77
0.51
0.53
0.45
0.48

0.86
0.47
0.44
0.50

0.91
0.43
0.43

0.85
0.46

0.92

Notes: 1. Diagonal elements (in shade) are the square root of the average variance extracted (AVE). Off-diagonal elements are the correlations among constructs. For
discriminant validity, diagonal elements should be larger than off-diagonal elements. 2. CON: conrmation; WQ: website quality; SAT: satisfaction; TRT: trust; HAB: habit;
RPT: repeat purchase intention.

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

53

supported. Finally, trust exerts stronger impact on repeat purchase

intention for customers with lower levels of habit, indicating H13
is supported.
6. Discussion
6.1. Key ndings

Fig. 2. SEM analysis of the research model.

purchase intention ( = 0.183, 0.175; t = 2.404; 3.071, respectively),

validating H9 and H10.
In order to further test the moderating effect of habit, we
employed the formula proposed by Chin, Marcolin, and Newsted
(1996) to evaluate the differences in path coefcients of two
models (i.e., high-habit subgroup and low-habit subgroup)1 by
calculating t-statistics, following Ahuja and Thatcher (2005) and
Chang et al. (2014). As shown in Appendix B, the relationship
between perceived value and repeat purchase intention is stronger
for customers with high levels of habit, while satisfaction has
higher inuence on repeat purchase intention for customers with
lower levels of habit. Thus H11 is supported, whereas H12 is not

Spooled = sqrt[(N1 1)/(N1 + N2 + 2)] SE12 + [(N2 1)/(N1 + N2 + 2)]

SE22
t = (PC1 PC2)/[spooled sqrt(1/N1 + 1/N2)]
where spooled = pooled estimator for the variance; t = t-statistic with N1 + N2 + 2
degrees of freedom; Ni = sample size of dataset for subgroup i; Sei = standard error
of path in structural model for subgroup i; Pci = path coefcient in structural model
of subgroup i.
1

The aim of this study is to integrate the literature of ECM

and online shopping to investigate the factors affecting customers repeat purchase intention. The results reveal that trust
and satisfaction are the two key factors affecting customers repeat
purchase intention. The ndings are consistent with prior studies (e.g., Liu et al., 2011; Shin et al., 2013; Tsao, 2013), provide
additional evidence to conrm the importance of relationship quality in the online shopping context. In addition, consistent with
Casal et al. (20102011) and Kim et al. (2004), the ndings
report that satisfaction has strong effect on trust, providing an
additional evidence to support the link between satisfaction and
trust.
We also nd that conrmation, website quality, and perceived
value maintain signicant relationships with customer satisfaction.
They are ranked in importance as follows: perceived value, conrmation, and website quality. The ndings indicate that perceived
value is the most important driver of customer satisfaction, similar
to the study of Chiu et al. (2012). The results indicate that conrmation exerts positive effects on perceived value and website quality,
which support the assertions of ECM and Fornell et al. (1996) that
conrmation and expectation will have positive inuence on customers perception of benets and quality. Finally, the results show
that website quality exerts signicant inuence on perceived value.
This nding supports the literature of D&M model (e.g., Wu & Wang,
2006) that website quality perception has positive inuence on
expected benets.
The ndings of this study reveal that habit will moderate the
impact of perceived value on repeat purchase intention. This nding extends the existing literature by addressing that habit exerts
stronger impact on repeat purchase intention for customer with
high levels of habit. Our results show that satisfaction has stronger
effect on repeat purchase intention in the low-habit subgroup.
However, this nding is contrary to the standpoint of Khalifa and
Liu (2007). We performed an additional PLS analysis for high-habit
subgroup, which indicated that the path coefcient of satisfaction
increased from 0.061 to 0.341 by removing perceived value from
the research model. One possible explanation for this nding might
be that when the impact of perceived value is taken into consideration, customers are more concerned the benets they can obtain.
Finally, the results also show that habit exerts stronger effect on the
link between trust and repeat purchase intention in the low-habit
subgroup. In line with the study of Chiu et al. (2012), this nding
indicates that strong habit may eliminate customers perception of
uncertainty and thus reduce the inuence of trust on behavioral
intention.
6.2. Implication for theory and practice
There are several implications for research and practice emerging from this study. First, although the roles of website quality,
value, and trust have been intensively examined by prior literature on e-commerce, little work has been done to combine website
quality, value, and trust to test their effects on repeat purchase
intention simultaneously and to examine the inuence of transaction experience (conrmation of expectations) on these constructs.
Therefore, developing a research model based on the literature of
ECM and online shopping to explain repeat purchase intention is

54

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

a contribution of this study. In addition, the results indicate that

online sellers should build appropriate level of customer expectations and do their best to meet those expectations to increase
customer satisfaction, value, and website quality. Since expectations may vary across different customers, a better strategy for
online sellers to identify customer expectations is to segment their
customers based on their needs and design different marketing
programs for each segment (Bhattacherjee, 2001a). This can be
done by using focus group method to identify customer expectations for each segment. In addition, since customers expectation
would be rened and modied based on their prior transaction
experiences (Bhattacherjee, 2001a), online sellers should modify their marketing programs continuously to meet customers
needs. Management of online group-buying could use social networking sites to communicate with customers to listen their
suggestions.
Second, although prior study on ECM has combined trust and
ECM to predict behavioral intention (Casal et al., 20102011), few
studies have been conducted to test the moderating role of habit
on the link between trust and behavioral intention. The ndings
of this study report that the relationship between repeat purchase
intention and trust is contingent upon online shopping habit. This
nding extends Casal et al.s (20102011) study by revealing that
the effect of trust on repeat purchase intention will reduce when
online shopping habit increases. From a practical standpoint, the
ndings imply that trust is an important factor in stimulating customers repeat purchase intention (Chiu et al., 2012). Once trust is
built and reaches a sufcient level, online sellers should encourage customers to visit website automatically, as suggested by Chiu
et al. (2012). Thus, online sellers could encourage frequent usage by
providing incentives (e.g., bonuses and prizes) and establish customer services sites to develop customers habit (Limayem et al.,
2007).
Third, although prior literature on ECM has integrated TAM
to examine the role of perceived ease of use in IS continuance intention (e.g., Hong et al., 2006), perceived ease of use in
TAM is limited in measure users perception of IS quality for a
new information technology such as online shopping websites,
because it was originally designed to explain users usage behavior of traditional technologies (e.g., e-mail and Spreadsheet) in
an organizational setting (Kim, Chan, & Gupta, 2007). Thus, by
integrating the aspect of website quality from D&M model into
ECM, this study extends existing literature by proving a more
complete viewpoint to explain the link between technical characteristics (e.g., system quality, information quality, and service
quality) and customer satisfaction. From a practical perspective,
the results mean that online sellers should improve technical
features such as system quality, service quality, and information
quality to increase customer satisfaction. This can be done by providing an information system with well-designed user interface,
high transmission quality, privacy protection mechanism, and safe
payment mechanism (Chiu et al., 2012; Li et al., 2006). Moreover, online sellers should encourage customers to enter their own
personal data to utilize personalized and customized services to
customers. By doing so, online sellers can interact with customers
to improve mutual understanding of customers and thus build
customers relational commitment, including satisfaction (Li et al.,
2006).
Furthermore, while perceived usefulness is a strong predictor of satisfaction and behavioral intention in the original ECM
model, previous views were incomplete. This is because perceived
usefulness is generally considered as a type of extrinsic benet
customers obtain from online transaction. In this sense, the role
of intrinsic benets has been ignored in ECM. This study theorizes and empirically shows that perceive value is a construct
consisting of both extrinsic (e.g., efciency and convenience of

online shopping) and intrinsic benets (e.g., enjoyment and hedonic) and is a strong predictor of customer satisfaction and repeat
purchase intention. The ndings mean that online sellers should
provide greater functional value and hedonic value for customers
to increase customer satisfaction and repeat purchase intention.
This can be done by providing an efcient and effective search
function and sufcient product information to help customers to
nd the desired products efciently and effectively (Chiu et al.,
2012).
Finally, by testing the moderating effects of habit on the relationships between repeat purchase and its antecedents, our results
report that satisfaction will predict lower level of repeat purchase intention for customer with high levels of habit, once
perceived value is taken into account. This nding provides
a possible explanation for the mixed results of Anderson and
Srinivasan (2003) and Khalifa and Liu (2007). That is, while the
effect of satisfaction on repeat purchase intention will become
stronger as habit of online shopping increases, habit will lower
the sensitivity of customer loyalty to satisfaction due to the suppressor effect of perceived value. Moreover, the results show that
perceived value is the most important predictor of repeat purchase intention. From the perspective of management of online
group-buying, enhancing customer value should be given higher
priority.

6.3. Limitations
Although our ndings provide several useful implications for
theory and practice, the study still faces several limitations. First,
the ndings of this study presented only a period of customers
feelings toward an online group-buying website. Factors inuencing repeat purchase intention might be different from other
types of website (e.g., the customer-imitated business model
websites). Second, the data are cross-sectional. Prior literature
argues that relational quality such as trust and satisfaction will
develop as interactions increase over time (e.g., Hsu et al., 2011).
Further studies should use longitudinal perspective to test the
inuences of trust and satisfaction on repeat purchase intention.
In addition, according to Hsu et al. (2014), trust and satisfaction in online group-buying can be divided into different types,
such as trust in sellers, trust in websites, satisfaction with sellers, and trust in websites. Further studies should be done to
test how habit moderates their effects on repeat purchase intention and to explore their antecedents as well. Third, the data
of this study were collected from customers of a group-buying
website in Taiwan. Future research should test the ndings of
our study in different countries (e.g., US) to test the generalizability of this study. Finally, there are some factors, such
as network externality (e.g., number of shopping partners and
number of peers) and price discount, may affect customers
repeat purchase intention. Further studies should be done to
explore the antecedents of repeat purchase intention from various
standpoints.

6.4. Conclusions
This study developed and tested a model explaining the determinants of repeat purchase intention. By including ECM, website
quality, perceived value, satisfaction, and habit, this study examined their effects on repeat purchase intention simultaneously and
tested the impact of conrmation of expectations on these determinants of repeat purchase intention. To a certain degree, the
ndings of this study provide useful implications for e-commerce
practitioners.

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

55

Appendix A. PLS conrmatory factor analysis and

cross-loadings.

CON1
CON2
CON3
PV1
PV2
PV3
PV4
WQ1
WQ2
WQ3
WQ4
WQ5
WQ6
WQ7
SAT1
SAT2
SAT3
TRT1
TRT2
TRT3
HAB1
HAB2
HAB3
RPI1
RPI2
RPI3

CON

PV

WQ

SAT

TRT

HAB

RPI

0.81
0.86
0.83
0.41
0.45
0.47
0.39
0.47
0.48
0.37
0.40
0.43
0.42
0.44
0.50
0.42
0.46
0.43
0.47
0.44
0.49
0.47
0.42
0.52
0.56
0.47

0.48
0.48
0.43
0.75
0.74
0.80
0.79
0.52
0.60
0.53
0.44
0.46
0.48
0.46
0.55
0.39
0.52
0.36
0.40
0.35
0.38
0.42
0.45
0.53
0.54
0.56

0.45
0.48
0.46
0.42
0.45
0.58
0.53
0.74
0.78
0.79
0.79
0.79
0.78
0.76
0.48
0.38
0.45
0.51
0.48
0.45
0.36
0.39
0.38
0.44
0.45
0.42

0.51
0.40
0.43
0.38
0.51
0.43
0.42
0.46
0.44
0.36
0.40
0.35
0.38
0.38
0.90
0.82
0.86
0.40
0.46
0.41
0.34
0.42
0.35
0.48
0.48
0.42

0.41
0.35
0.47
0.28
0.30
0.34
0.33
0.40
0.40
0.39
0.42
0.46
0.43
0.38
0.46
0.37
0.37
0.89
0.93
0.89
0.32
0.36
0.40
0.40
0.42
0.37

0.45
0.46
0.43
0.33
0.40
0.42
0.36
0.42
0.39
0.33
0.33
0.33
0.32
0.28
0.41
0.29
0.41
0.38
0.42
0.36
0.85
0.87
0.84
0.45
0.47
0.34

0.47
0.51
0.43
0.42
0.48
0.50
0.43
0.44
0.38
0.40
0.33
0.37
0.34
0.32
0.52
0.31
0.45
0.39
0.41
0.37
0.35
0.40
0.42
0.94
0.94
0.86

Notes: 1. Bold numbers indicate item loadings on the assigned constructs. 2. CON: conrmation; WQ; website quality; SAT: satisfaction; TRT: trust; HAB: habit; RPT: repeat
purchase intention.

Appendix B. Statistical comparison of paths

Paths

Perceived value continuance intention

Satisfaction continuance intention
Trust continuance intention
*
**
***

High-habit (R2 = 0.354)

Low-habit (R2 = 0.357)

Statistical comparison
of paths

Standardized path coefcient

T-value

Standardized path coefcient

T-value

0.541
0.061
0.070

6.071***
0.510
0.905

0.330
0.241
0.175

2.466*
3.022**
0.281**

5.283***
4.307***
3.005**

p < 0.05.
p < 0.01.
p < 0.001.

References
Ahuja, M. J., & Thatcher, J. B. (2005). Moving beyond intention and toward the theory
of trying: Effects of work environment and gender on post-adoption information
technology use. MIS Quarterly, 29(3), 427459.
Anderson, R. E., & Srinivasan, S. S. (2003). E-satisfaction and E-loyalty: A contingency
framework. Psychology & Marketing, 20(2), 123138.
Armstrong, J. S., & Overton, T. (1976). Estimating nonresponse bias in mail surveys.
Journal of Marketing Research, 14, 396402.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction
in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 11731182.
Bhattacherjee, A. (2001b). An empirical analysis of the antecedents of electronic
commerce service continuance. Decision Support Systems, 32(2), 201214.
Bhattacherjee, A. (2001a). Understanding information systems continuance: An
expectationconrmation model. MIS Quarterly, 25(3), 351370.
Blau, P. M. (1964). Exchange and power in social life. NY: Wiley.
Business-Next. (2012). Market share of Groupon Taiwan on May, 2012. Business Next
Publishing Corp. http://www.bnext.com.tw/focus/view/cid/103/id/23486
Casal, L. V., Flavin, C., & Guinalu, M. (20102011). Antecedents and consequences
of consumer participation in on-line communities: The case of the travel sector.
International Journal of Electronic Commerce, 15(2), 137167.
Chang, C. M., Hsu, M. H., Hsu, C. S., & Cheng, H. L. (2014). Examining the role
of perceived value in virtual communities continuance: Its antecedents and
the inuence of experience. Behaviour & Information Technology, 33(5), 502521.
Chea, S., & Luo, M. M. (2008). Post-adoption behaviors of e-service customers: The
interplay of cognition and emotion. International Journal of Electronic Commerce,
12(3), 2956.
Chen, J., Chen, X., & Song, X. (2007). Comparison of the group-buying auction and the xed pricing mechanism. Decision Support Systems, 43(2),
445459.

Cheng, H.-H., & Huang, S.-W. (2013). Exploring antecedents and consequence of
online group-buying intention: An extended perspective on theory of planned
behavior. International Journal of Information Management, 33, 185198.
Chin, W. W., Marcolin, B. L., & Newsted, P. R. (1996). A partial least squares latent
variable modeling approach for measuring interaction effects: Results from a
Monte Carlo Simulation study and voice mail emotion/adoption study. In J. I.
DeGross, S. Jarvenpaa, & A. Srinivasan (Eds.), Proceedings of the 17th international
conference on information systems Cleveland, OH, (pp. 2141). Available from
http://disc-nt.cba.uh.edu/chin/icis96.pdf. Accessed 11.12.12
Chiu, C. M., Hsu, M. H., Lai, H., & Chang, C. M. (2012). Re-examining the inuence of
trust on online repeat purchase intention: The moderating role of habit and its
antecedents. Decision Support Systems, 53(4), 835845.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance
of information technology. MIS Quarterly, 13(3), 319340.
DeLone, W. H., & McLean, E. R. (2003). The DeLone and McLean model of information
systems success: A ten-year update. Journal of Management Information Systems,
19(4), 930.
DeLone, W. H., & McLean, E. R. (2004). Measuring e-commerce success: Applying
the DeLone & McLean information systems success model. International Journal
of Electronic Commerce, 9(1), 3147.
Doney, P. M., & Cannon, J. P. (1997). An examination of the nature of trust in
buyerseller relationships. Journal of Marketing, 3551.
Devaraj, S., Fan, M., & Kohli, R. (2002). Antecedents of B2C channel satisfaction and
preference: Validating e-commerce metrics. Information Systems Research, 13(3),
316333.
Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson and
Company.
Fornell, C., Johnson, M. D., Anderson, E. W., Cha, J., & Bryant, B. E. (1996). The American
customer satisfaction index: Nature, purpose, and ndings. Journal of Marketing,
718.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 3950.

56

M.-H. Hsu et al. / International Journal of Information Management 35 (2015) 4556

Gefen, D., Karahanna, E., & Straub, D. W. (2003). Trust and TAM in online shopping:
An integrated model. MIS Quarterly, 27(1), 5190.
Gupta, S., & Kim, H. W. (2007). The moderating effect of transaction experience on
the decision calculus in on-line repurchase. International Journal of Electronic
Commerce, 12(1), 127158.
Gefen, D., Straub, D. W., & Boudreau, M. C. (2000). Structural equation modeling and
regression: Guidelines for research practice. Communications of the association
for information systems, 4(7), 176.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate
data analysis (6th ed.). Upper Saddle River, NJ: Prentice Hall.
Hong, S., Thong, J. Y. L., & Tam, K. Y. (2006). Understanding continued information
technology usage behavior: A comparison of three models in the context of
mobile Internet. Decision Support Systems, 42, 18191834.
Hsu, M. H., Chang, C. M., Chu, K. K., & Lee, Y. J. (2014). Determinants of repurchase intention in online group-buying: The perspectives of DeLone & McLean
IS success model and trust. Computers in Human Behavior, 36, 234245.
Hsu, M. H., Chang, C. M., & Yen, C. H. (2011). Exploring the antecedents of
trust in virtual communities. Behaviour & Information Technology, 30(5), 587
601.
iThome. (2011). More than 20% of Internet users purchased products in the online groupwebsites.
http://www.ithome.com.tw/itadm/article.php?c=65725,.
buying
Accessed 10.09.13
Kang, Y. S., Hong, S., & Lee, H. (2009). Exploring continued online service usage
behavior: The roles of self-image congruity and regret. Computers in Human
Behavior, 25, 111122.
Kauffman, R. J., Lai, H., & Ho, C. T. (2010). Incentive mechanisms, fairness and participation in online group-buying auctions. Electronic Commerce Research and
Applications, 9, 249262.
Kauffman, R. J., Lai, H. C., & Lin, H. C. (2010). Consumer adoption of group-buying
auctions: An experimental study. Information Technology and Management, 11,
191211.
Kauffman, R. J., & Wang, B. (2002). Big together, buy together: On the efcacy of
group-buying business models in Internet-based selling. In P. B. Lowry, J. O.
Cherrington, & R. R. Watson (Eds.), Handbook of Electronic Commerce in Business
and Society (pp. 99137). Boca Raton, FL: CRC Press.
Khalifa, M., & Liu, V. (2007). Online consumer retention: Contingent effects of online
shopping habit and online shopping experience. European Journal of Information
Systems, 16(6), 780792.
Kim, H. W., Chan, H. C., & Gupta, S. (2007). Value-based adoption of mobile Internet:
An empirical investigation. Decision Support Systems, 43(1), 111126.
Kim, H. W., & Gupta, S. (2009). A comparison of purchase decision calculus between
potential and repeat customers of an online store. Decision Support Systems, 47,
477487.
Kim, H., Suh, K. S., & Lee, U. K. (2013). Effects of collaborative online shopping on
shopping experience through social and relational perspectives. Information &
Management, 50, 169180.
Kim, H. W., Xu, Y., & Koh, J. (2004). A comparison of online trust building factors
between potential customers and repeat customers. Journal of the Association
for Information Systems, 5(10), 392420.
Lee, M.-C. (2010). Explaining and predicting users continuance intention toward
e-learning: An extension of the expectationconrmation model. Computers &
Education, 54(2), 506516.
Lee, Y., & Kwon, O. (2011). Intimacy, familiarity and continuance intention: An
extended expectationconrmation model in web-based services. Electronic
Commerce Research and Applications, 10(3), 342357.
Li, D., Browne, G. J., & Wetherbe, J. C. (2006). Why do Internet users stick with a
specic web site? A relationship perspective. International journal of electronic
commerce, 10(4), 105141.
Liang, H., Saraf, N., Hu, Q., & Xue, Y. (2007). Assimilation of enterprise systems: The
effect of institutional pressures and the mediating role of top management. MIS
Quarterly, 31(1), 5987.
Limayem, M., Hirt, S. G., & Cheung, C. M. K. (2007). How habit limits the predictive
power of intention: The case of information systems continuance. Management
Information Systems Quarterly, 31(4), 705737.
Lin, H. H., & Wang, Y. S. (2006). An examination of the determinants of customer loyalty in mobile commerce contexts. Information & Management, 43(3), 271282.
Lin, C. S., Wu, S., & Tsai, R. J. (2005). Integrating perceived playfulness into
expectationconrmation model for web portal context. Information & Management, 42(5), 683693.
Lin, T. C., Wu, S., Hsu, J. S. C., & Chou, Y. C. (2012). The integration of value-based adoption and expectationconrmation models: An example of IPTV continuance
intention. Decision Support Systems, 54, 6375.
Liu, C. T., Guo, Y. M., & Lee, C. H. (2011). The effects of relationship quality and
switching barriers on customer loyalty. International Journal of Information Management, 31(1), 7179.
Liu, M. T., Brock, J. L., Shi, G. C., Chu, R., & Tseng, T. H. (2013). Perceived benets,
perceived risk, and trust: Inuences on consumers group buying behaviour.
Asia Pacic Journal of Marketing and Logistics, 25(2), 225248.

Market Intelligence & Consulting Institute. (2011). The customer behavior

of group-buying auctions sites in 2010. Institute for Information Indushttp://www.mic.iii.org.tw/aisp/pressroom/press01 pop.asp?sno=243
try.
&cred=2011/1/25&type1=2,. Accessed 27.04.13
Oliver, R. L. (1992). An investigation of the attribute basis of emotion and related
affects in consumption: Suggestions for a stage-specic satisfaction framework.
Advances in Consumer Research, 19(1), 237244.
Ouellete, J. A., & Wood, W. (1998). Habit and intention in everyday life: The multiple
processes by which past behavior predicts future behavior. Psychological Bulletin,
124(1), 5474.
Pavlou, P. A., & Fygenson, M. (2006). Understanding and predicting electronic commerce adoption: An extension of the theory of planned behavior. MIS quarterly,
30(1), 115143.
Pavlou, P. A., Liang, H., & Xue, Y. (2007). Understanding and mitigating uncertainty
in online exchange relationships: A principalagent perspective. MIS quarterly,
31(1), 105136.
Pennington, R., Wilcox, H. D., & Grover, V. (20032004). The role of system trust in
business-to-consumer transactions. Journal of Management Information Systems,
20(3), 197226.
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature
and recommended remedies. Journal of applied psychology, 88(5), 879
903.
Qureshi, I., Fang, Y., Ramsey, E., McCole, P., Ibbotson, P., & Compeau, D. (2009). Understanding online customer repurchasing intention and the mediating role of trust
An empirical investigation in two developed countries. European Journal of
Information Systems, 18(3), 205222.
Recker, J. (2010). Explaining usage of process modeling grammars: Comparing three
theoretical models in the study of two grammars. Information & Management,
47(5), 316324.
Ringle, C. M., Wende, S., & Will, S. (2005). SmartPLS 2.0 (M3) Beta. Hamburg.
http://www.smartpls.de
Shin, J. I., Chung, K. H., Oh, J. S., & Lee, C. W. (2013). The effect of site quality on
repurchase intention in Internet shopping through mediating variables: The
case of university students in South Korea. International Journal of Information
Management, 33(3), 453463.
Teo, T. S. H., Srivastava, S. C., & Jiang, L. (20082009). Trust and electronic government
success: An empirical study. Journal of Management Information Systems, 25(3),
99132.
Tsao, W.-Y. (2013). The tness of product information: Evidence from online recommendations. International Journal of Information Management, 33(1), 19.
Udo, G. J., Bagchi, K. K., & Kirs, P. J. (2010). An assessment of customers e-service
quality perception, satisfaction and intention. International Journal of Information
Management, 30, 481492.
Verplanken, B., Aarts, H., & van Knippenberg, A. (1997). Habit, information acquisition, and the process of making travel mode choices. European Journal of Social
Psychology, 27(5), 539560.
Wu, J. H., & Wang, Y. M. (2006). Measuring KMS success: A respecication of the
DeLone and McLeans model. Information & Management, 43(6), 728739.
Wood, W., & Neal, D. T. (2007). A new look at habits and the habit-goal interface.
Psychological review, 114(4), 843863.
Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought,
emotion, and action. Journal of Personality and Social Psychology, 83(6),
12811297.
Zeithaml, V. A., Parasuraman, A., & Malhotra, A. (2002). Service quality delivery
through web sites: A critical review of extant knowledge. Journal of the Academy
of Marketing Science, 30(4), 362375.
Meng-Hsiang Hsu holds a Ph.D. degree from National Sun-Yat-Sen University,
Taiwan. He is currently a Professor at the Department of Information Management,
National Kaohsiung First University of Science and Technology. His research interests include knowledge management, information ethics, and electronic commerce.
He has published articles in Decision Support Systems, Information Systems Journal, Journal of Business Ethics, Behaviour & Information Technology, International
Journal of HumanComputer Studies, and others.
Chun-Ming Chang is currently an associate professor at the Department of Tourism
Information, Aletheia University, Taiwan. He holds a Ph.D. degree in Management
from National Kaohsiung First University of Science and Technology, Taiwan. His
current research interests include electronic commerce and knowledge management. He has published articles in Decision Support Systems, Information Systems
Journal, International Journal of HumanComputer Studies, Journal of Knowledge
management, and Behaviour & Information Technology, and others.
Li-Wen Chuang is a Assistant professor at Fuzhou University, P.R.O.C. He conducts
research in e-business, social media, and information literacy. He has published articles in Internet Research, American Journal of Engineering and Technology Research,
Applied Mechanics and Materials, and others.