Article

Information Development
2016, Vol. 32(1) 2034
An investigation of users The Author(s) 2014
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
continuance intention towards DOI: 10.1177/0266666914522140
idv.sagepub.com
mobile banking in China

Shunbo Yuan, Yong Liu, Ruihong Yao

Jiaxing University

Jing Liu
Nanjing University of Posts and Telecommunications

Abstract
The long-term development of mobile banking (m-banking) relies on users continued usage. Motivated by the
need to better understand the motivations and barriers of users continuance intention towards m-banking,
this study develops a research model based on the incorporation of the technology acceptance model (TAM):
task-technology fit model (TTF) and perceived risk into the expectance-confirmation model (ECM). Empirical
data from 434 users who had prior experience with m-banking were tested against the proposed research
model by using structural equation modeling (SEM). The results indicate that satisfaction, perceived usefulness,
perceived task-technology fit, and perceived risk are the main predictors of continuance intention, satisfaction,
in turn, is determined by confirmation, perceived usefulness, and perceived risk. Perceived usefulness is
affected by confirmation, perceived ease of use, and perceived task-technology fit. However, the direct effect
of perceived ease of use to continuance intention is not significant. The results also show that gender signifi-
cantly moderates the effect of perceived risk to continuance intention. Implications of the findings and future
research directions are discussed.

Keywords
mobile commerce, mobile banking, continuance intention, expectance-confirmation model, technology
acceptance model, task-technology fit, China

Monitoring and improving users satisfaction with m-banking is an appropriate approach

for providers to retain users.

Introduction the number of m-banking users will exceed 590 mil-

Mobile banking (m-banking) means that users adopt lion in 2013 and more than 1 billion mobile phone
mobile terminals such as cell phones to access pay- users will have used their phones for m-banking by
ment services including account inquiry, transference the end of 2017 (Morris, 2013). In China, four state-
and bill payment (Dahlberg et al., 2008). Compared to owned banks, including the Bank of China (BOC):
traditional/online banking, the main advantages of Industrial and Commercial Bank of China (ICBC):
m-banking are ubiquity and immediacy. That is, Agricultural Bank of China (ABC): and China Con-
m-banking can free users from temporal and spatial struction Bank (CCB): along with a number of private
limitations, and enable them to conduct payment at
any time from anywhere (Zhou, 2012).
Corresponding author:
Thanks to the increasingly wide spread of mobile
Shunbo Yuan, School of Business, Jiaxing University, 56 South
communication technologies represented by 3G net- Yuexiu Rd., Jiaxing, Zhejiang Province, P.R. China. Phone: 86
works, m-banking has gained rapid growth world- 15888317156.
wide. According to a report from Juniper Research, Email: alexshun915@gmail.com
Yuan et al: An investigation of users continuance intention towards mobile banking in China 21

banks such as China Merchant Bank (CMB): China Third, the ECM and TAM only explain the continu-
Minsheng Banking Corp., Ltd., and Industrial Bank ance intention from the perspective of users percep-
Co., Ltd., have released their own m-banking ser- tions towards technology, while users continuance
vices. The number of m-banking users of three banks intention is not only determined by their perceptions
(ABC, BOC, and CCB) had reached 180 million by and attitudes, but also by task-technology fit. To fill
the end of 2012 (CEBNET, 2013). this gap, our study combines TTF with TAM and ECM
The rapid growth of m-banking has presented not to further our understandings of the determinants of
only opportunities, but also challenges, to the service m-banking continuance intention. Fourth, adding per-
providers. As Bhattacherjee (2001a) argued, long- ceived risk to the model allows us to capture the possi-
term viability of an information system (IS) and its ble element of barriers to the m-banking continuance
eventual success depend on its continued use rather intention.
than first-time use. To m-banking providers, retain- The remainder of the article proceeds as follows.
ing their existing users is a crucial source of compet- The literature review and theoretical background of
itive advantage. They have invested great effort and this study are described in the next section. The third
resources on releasing m-banking services; only section introduces our research model and hypoth-
retaining users and facilitating users continuance eses. Research methodology is presented in the fourth
usage can recover these costs and achieve success. section. The results are described in the fifth Section,
If users enthusiasm over the initial adoption followed by a discussion in the sixth section. The arti-
diminishes after gaining usage experience, the cle concludes with implications and directions for
m-banking services will suffer from decreased usage future studies.
and may even fall into disuse. When this happens,
the service providers might need to write off their
significant investments. Thus, continuance intention Literature review and theoretical
has become an important subject of study in the background
m-banking research area. However, compared to the
abundant research on initial adoption, the continu- Users adoption of m-banking
ance intention of m-banking has seldom been exam- Extant research has focused on examining the factors
ined, which involves great uncertainty and risk that affecting users initial adoption of m-banking (Brown
may inhibit users continuance usage. To address et al., 2003; Laforet and Li, 2005; Laukkanen, 2007;
this gap, conducting an investigation of the factors Yang, 2009; Puschel et al., 2010; Dasgupta et al.,
affecting users continuance intention is necessary. 2011; Zhou, 2011a; Zhou, 2012; Chen, 2013). Infor-
The main purpose of this study is to enhance our mation technology adoption theories such as TAM
understanding of the formation of users continuance (Luarn and Lin, 2005; Amin et al., 2008; Dasgupta
intention towards m-banking and to offer a helpful et al., 2011): innovation diffusion theory (IDT) (Suor-
way for m-banking service providers to improve anta and Mattila, 2004; Riquelme and Rios, 2010;
user retention. Hence, we incorporate the technology Koenig-Lewis et al., 2010; AI-Jabri and Sohail,
acceptance model (TAM): the task-technology fit 2012): (decomposed) theory of planned behavior
model (TTF) and perceived risk into the expectance- (DTPB/TPB) (Brown et al., 2003; Riquelme and Rios,
confirmation model (ECM) to hypothesize a new 2010; Puschel et al., 2010; Sripalawat et al., 2011):
model and conduct an empirical study. We make this and unified theory of acceptance and use of technol-
incorporation for the following four reasons. First, ogy (UTAUT) (Tan et al., 2010; Zhou et al., 2010;
although the ECM has been confirmed by previous Yu, 2012) are often used as the theoretical bases. Per-
studies to be a robust model for explaining continuance ceived usefulness, perceived ease of use, relative
intention of IS, it employs only three predictors, advantage, subjective norms, self-efficacy, perfor-
namely satisfaction, perceive usefulness, and confir- mance expectancy, and facilitating conditions are
mation, to explain continuance intention. However, identified to affect users adoption. In addition, per-
users behavior intention towards IS may be affected ceived risk (Cruz et al., 2010; Chen, 2013): initial
by other factors. Second, TAM provides an important trust (Zhou, 2011a): trust (Koo and Wati, 2010): flow
belief, namely perceived ease of use, as the critical fac- experience (Zhou, 2012): innovation attributes and
tor affecting perceived usefulness, which makes up the knowledge-based trust (Lin, 2011) are also validated
precedent factor of perceived usefulness for ECM. to have positive/negative effects.
22 Information Development 32(1)

Continuance intention of mobile services usefulness to be the surrogate for post-adoption

Compared to the abundant research on initial adop- expectation, which can make up for the shortcomings
tion, continuance intention has received relatively of ECT (Thong et al., 2006). Many studies have used
less attention. Access quality, usefulness, uncertainty the ECM to understand users continued usage of IS
avoidance, perceived enjoyment and familiarity were such as online banking (Vatanasombut et al., 2008),
verified to affect the continuance intention of mobile e-learning (Lee, 2010), Internet-based learning tech-
Internet services (Lee et al., 2007; Ng and Kwahk, nologies (Limayem and Cheung, 2008), online shop-
2010; Shin et al., 2010; Zhou, 2011b). Similarly, some ping (Lee and Kwon, 2011), and IPTV (Lin et al.,
researchers noted that information quality, social influ- 2012), which show that the ECM is effective in pre-
ence, perceived enjoyment and perceived fee have dicting continuance intention in the context of mobile
effects on continuance intention of mobile data ser- commerce. Therefore, the ECM is employed to under-
vices (Kim et al., 2008; Lee et al., 2009; Kim, stand users continuance intention towards m-banking
2010; Choi et al., 2011). in the current study.
Chen (2012) suggested that technology readiness
and service quality have indirectly significant influ- Technology acceptance model (TAM)
ence on continuance intention of m-banking. Chen Based on the theory of reasoned action (TRA), Davis
and Liu (2012) argued that the continuous improve- proposed the TAM and aimed to develop a model for
ment of mobile content services has a positive impact understanding users acceptance of IS (Davis, 1989).
on users expectations. Kang et al. (2012) noted that As the TAM indicates, the actual use of IS is deter-
perceived usability, channel preference, and per- mined by users intention. The intention, in turn, is
ceived value are three major determinants of sus- determined by users attitude towards using an IS and
tained m-banking use. Zhao et al. (2012) explored its perceived usefulness. Perceived usefulness and
the effects of service quality and justice on user satis- perceived ease of use are the determinants of attitude
faction, which in turn, affects continuance intention of towards IS. In addition, perceived ease of use affects
mobile value-added services. Zhou (2013) drew on perceived usefulness, which also mediates the effects
the information systems success model and flow the- of perceived ease of use on intention (Davis, 1989;
ory to examine the effects of extrinsic and intrinsic Davis et al., 1989).
factors on continuance intention of mobile payment TAM was originally developed to predict users
services. Hsiao and Chang (2013) found that per- initial adoption of a new IS. Many studies have con-
ceived value, perceived usefulness, and satisfaction firmed that TAM can efficiently predict and explain
all directly influence the continuance intention of users behaviors in using IS (Moon and Kim, 2001;
mobile advertising. Shin, 2007; Park et al., 2012). Meanwhile, the TAM
has been implemented to investigate users continu-
Expectance-Confirmation Model (ECM) ance intention of IS by several studies (Gefen, 2003;
Chan and Lu, 2004; Kim and Malhotra, 2005; Hong
Based on expectation-confirmation theory (ECT): a
et al., 2006; Chang et al., 2013). The two main con-
theory widely used in the consumer behavior litera-
structs, namely perceived usefulness and perceived
ture to study consumer satisfaction, post-purchase
ease of use, have also been extensively validated to
and service marketing in general, Bhattacherjee pro-
be the antecedent factors affecting users continuance
posed the Expectance-Confirmation Model (ECM).
intention. Since TAM is a very parsimonious and gen-
Bhattacherjee suggested that satisfaction with IS use
eric model which can be used to study both initial and
is the strongest predictor of users continuance inten-
continued IS adoption (Hong et al., 2006), we incor-
tion, followed by perceived usefulness as a significant
porate perceived ease of use into the basic framework
but weaker predictor. Confirmation and perceived
of ECM to better explain users continuance intention.
usefulness are major determinants of satisfaction
(Bhattacherjee, 2001a).
As a theoretical model focusing on the continuance Task-Technology Fit (TTF)
usage of IS, the ECM can help us to understand user TTF is a very precise exploration of the links between
behaviors that cannot be explained by the initial adop- work-related matters and the use of technology pro-
tion theories. In addition, the ECM posits that users posed by Goodhue and Thompson (1995). Compared
expectations will change after adoption and perceived with the TAM, TTF focuses more on the relationships
Yuan et al: An investigation of users continuance intention towards mobile banking in China 23

among task and technology characteristics, utilization The level of satisfaction is affected by the level of
and performance impact. The theory holds that the confirmation.
degree of fit between tasks and technology will The post-adoption expectations regarding the IS,
impact work performance and technology utilization. another determinant of satisfaction, are often repre-
IS are more likely to have a positive impact on indi- sented by perceived usefulness. Perceived usefulness
vidual performance and be used if the capabilities of is an important predictor of behavioral intention in
the IS match the tasks that the user must perform. TTF many contexts including the Internet, information
is a powerful model to analyze users adoption and systems, the web, and mobile commerce. The ECM
behaviors of IS in a specific context, such as mobile posits that perceived usefulness has a positive effect
commerce in the insurance industry (Lee et al., on users satisfaction and continuance intention. In
2007), mobile information systems (Junglas et al., addition, users confirmation of expectations will
2008): e-learning (Larsen et al., 2009; Lin, 2012), and have a positive effect on the perceived usefulness of
e-books (DAmbra et al., 2013). Since the TTF shows IS (Bhattacherjee, 2001a, 2001b). Perceived useful-
a good complementarity with the TAM and ECM, this ness of IS can be adjusted by confirmation experi-
study incorporates the perceived task-technology fit ence, particularly when the users initial perceived
into the ECMs framework. usefulness is not concrete due to the uncertainty over
what to expect from using the IS (Lee, 2010). Thus,
Research model and hypotheses the following hypotheses are proposed:
As the ECM employs only three predictors to explain
H2: Perceived usefulness is positively related to
continuance intention, while users behavior may be
users continuance intention of m-banking.
affected by other factors, it is appropriate to incorpo-
rate other models into the ECM. TAM and TTF have H3: Perceived usefulness is positively related to
been used by numerous studies to examine users users satisfaction with m-banking.
adoption of IS and have good complementarities to H4: Confirmation of expectations is positively
the ECM. Consequently, we develop our research related to users satisfaction with m-banking.
model and hypotheses based on the incorporation of H5: Confirmation of expectations is positively
TAM, TTF, and perceived risk into the ECM. There related to perceived usefulness of m-banking.
are five main hypotheses in the ECM. First, continu-
ance intention is directly affected by the users satis- Davis (1989) defined perceived ease of use as the
faction with IS. Satisfaction refers to an overall degree to which a person believes that using a partic-
evaluation of an IS, which reflects an emotion-based ular system would be free of effort. In the TAM, per-
response about the target IS (Lam et al., 2004; Kim, ceived ease of use is assumed to indirectly affect
2010). Numerous studies pertaining to IS continuance behavioral intention by impacting perceived useful-
support that satisfaction is a major factor determining ness. The empirical evidence accumulated over two
the continuance intention. It is expected that continu- decades reveals that perceived ease of use is signifi-
ance intention of m-banking is positively determined cantly linked to intention, both directly and indir-
by users satisfaction. ectly, via its effect on perceived usefulness (Kim and
Malhotra, 2005; Shin, 2007; Chang et al., 2013). Com-
H1: Users satisfaction with m-banking is posi-
pared with computers, mobile terminals suffer some
tively related to their continuance intention of
constraints, such as small screens and inconvenient
m-banking.
input. If m-banking service providers cannot present
According to the ECM, post-adoption expectations a good interface, conducting m-banking transactions
regarding the IS and discrepancies between pre- will not be an easy task for users, which will not be
adoption expectations and actual performance of the favorable to their willingness to continued use of
IS are the two determinants of users satisfaction. m-banking. An ease-of-use m-banking system will
When users actually use m-banking, they update their reduce users effort spent and increase their feelings
initial expectations simultaneously. If m-banking out- of perceived usefulness. Therefore, we propose:
performs relative to their initial expectations, their
post-adoption expectations are confirmed. Otherwise, H6: Perceived ease of use is positively related to
their post-adoption expectations are not confirmed. users continuance intention of m-banking.
24 Information Development 32(1)

H7: Perceived ease of use is positively related to (2012) reported that perceived risk has no significant
perceived usefulness of m-banking. impact on the relationship quality of m-banking,
and Kang et al. (2012) found that perceived risk shows
TTF suggests that the degree of task-technology fit is a no significant effect on perceived value of m-banking.
major factor affecting users adoption of technology. Base on the previous discussions, we assume that per-
Goodhue and Thompson (1995) argued that a technology ceived risk has a negative effect on continuance inten-
will be adopted if it is . . . a good fit with the task it sup- tion, both directly and indirectly, via satisfaction with
ports. The extant research has also found significant m-banking.
correlations between task-technology fit and continu-
ance intention (Larsen et al., 2009; Lin, 2012). In the H11: Perceived risk is negatively related to us-
m-banking context, a good task-technology fit will ers continuance intention of m-banking.
promote users continuance intention, while a poor H12: Perceived risk is negatively related to us-
task-technology fit will decrease the intention. One ers satisfaction with m-banking.
of the most significant advantages of m-banking is
that users can enjoy m-banking services with their The moderating effects of gender in online/mobile
smartphones at any time and anywhere. If users do service settings have been discussed in previous stud-
need to conduct payment or other types of banking ies. Garbarino and Strahilevitz (2004) found women
transactions ubiquitously, the level of satisfaction perceive more risk than men in online purchases.
will be improved and their continuance intention Nysveen et al. (2005) found that peer opinion has a
will be more intensive. Otherwise, they may possi- higher effect on females in mobile services. Riquelme
bly abandon their usage and select traditional or and Rios (2010) and Yu (2012) suggested that gender
online banking services. In addition, the correlations plays a moderating role in the adoption of m-banking.
between perceived task-technology fit and perceived In contrast, no research has been conducted to inves-
usefulness have been confirmed by previous studies tigate the role of gender in the m-banking continued
(Dishaw and Strong, 1999; Zhou et al., 2010). Only usage context. Therefore, we experiment to propose
when the advantages of m-banking meet the users the hypotheses that the effects of antecedents to con-
requirements of banking transactions will they feel tinuance intention are moderated by gender.
that m-banking is useful. Thus, we assume:
H13a: The effect of satisfaction to continuance
H8: Perceived task-technology fit is posi- intention is moderated by gender.
tively related to users continuance intention of H13b: The effect of perceived usefulness to con-
m-banking. tinuance intention is moderated by gender.
H9: Perceived task-technology fit is positively H13c: The effect of perceived ease of use to
related to users satisfaction with m-banking. continuance intention is moderated by gender.
H10: Perceived task-technology fit is positively H13d: The effect of perceived task-technology fit
related to perceived usefulness of m-banking. to continuance intention is moderated by gender.
H13e: The effect of perceived risk to continu-
Perceived risk is one of the important components of ance intention is moderated by gender.
several IS adoption models. It reflects users percep-
tions of the uncertainty and adverse consequences of Figure 1 depicts our theoretical model of users
engaging in an activity. Tan et al. (2010) claimed that continuance intention towards m-banking, which
insufficient or unreliable security technologies will includes all the hypotheses mentioned above.
increase users perception of risk, resulting in lower
level satisfaction and willingness to continued usage Research methodology
in an e-commerce context. Wu and Wang (2005) found
that perceived risk has a significant direct effect on Questionnaire development
behavioral intentions in e-commerce. In terms of A questionnaire survey with two parts was used to test
m-banking, some studies claimed that perceived risk our theoretical model. The first part comprised a set of
is one of the main determinants of users adoption questions measuring the constructs in the research
(Brown et al., 2003; Luarn and Lin, 2005). While some model, and the second part concentrated on obtaining
other studies reached an opposite conclusion, Chen demographic information of the sample. Most items
Yuan et al: An investigation of users continuance intention towards mobile banking in China 25

Figure 1. Research model and hypotheses.

for constructs were drawn from pre-validated studies number of factors and whether the individual items
and reworded to fit the m-banking environment. The were loaded on their appropriate factors (Lee, 2010).
scale items for perceived usefulness, satisfaction, con- Each item in our study has high loading (> 0.5) on its
firmation, and continuance intention were drawn from related factors and low cross-loading (< 0.4) on other
Bhattacherjee (2001a, 2001b), while the scale items factors, indicating that the questionnaire has good con-
for perceived ease of use were adapted from Davis vergent and discriminate validities.
(1989). In addition, the items measuring perceived
task-technology fit were drawn from Goodhue and
Data collection and sample
Thompson (1995), and the perceived risk items were
adapted from Kang et al. (2012). The final items and Our study employed a survey of m-banking users to
their sources are listed in the Appendix A. All items gather data for hypotheses testing and to address research
corresponding to the constructs were measured by objectives. Before conducting the survey, we first con-
using a 7-point Likert-type scales anchored from 1 sidered the sample size. Stevens (2002) suggested that
(strongly disagree) to 7 (strongly agree). the sample size for social science research should be
Before conducting the survey, the questionnaire greater than 15 times the number of predictors. There are
was first sent to six experts and users of mobile com- 6 predictors in our model and the sample size should
merce to identify problems in the wording, content, therefore be larger than 90. In addition, as we needed
clarity, and quantities of items, and was then revised to use Structural Equation Modeling (SEM) to test the
according to their comments and suggestions. hypotheses, a sample size of at least 200 was necessary
(Hoetler, 1983). Thus, we set the minimum sample size
at 200, collected from nine of the three largest telecom-
Pilot test munication service providers in China (China Mobile,
The revised questionnaire was tested by convenience China Unicom, and China Telecom), for there are plenty
sampling. A total of 137 responses were received, of of respondents at these places and we could obtain a large
which 111 were valid, accounting for a valid response sample size conveniently. They are located in three cities
rate of 81 percent using Cronbachs reliability and of two Chinese eastern provinces. We think our survey
Exploratory Factor Analysis (EFA). The reliability method is appropriate, as eastern China is the area with
coefficients were first calculated for the items of each more mature mobile commerce than other parts of China.
construct, and the standard lower bound for Cronbachs We recruited respondents randomly in the mornings and
alpha was set at 0.7 (Anderson and Gerbing, 1988), afternoons during 10 weekdays and 2 weekends in
with items that did not significantly contribute to relia- August 2013, to remove potential sampling bias. The
bility being eliminated. EFA was then performed to users who had m-banking experience were asked to fill
examine whether the items produced the anticipated out the questionnaire voluntarily based on their usage
26 Information Development 32(1)

Table 1. Demographic information of the sample. Measurement model

Gender Male 39.6% The validity of the measurement model was evaluated by
Female 60.4% investigating convergent validity and discriminant valid-
Age under 20 5.8% ity. Convergent validity indicates the extent to which the
2030 24.4% items of a scale that are theoretically related should cor-
3040 36.2% relate highly. Composite Reliability (CR) and Average
4050 19.6% Variance Extracted (AVE) are the two most common
over 50 14.0% indices for convergent validity of measures (Lee et al.,
Occupation Students 13.8% 2007), and a CR of 0.70 or above is deemed acceptable
Clerical workers 69.6%
(Fornell and Larcker, 1987). AVE reflects the overall
Housewife 6.9%
amount of variance in the indicators accounted for by the
Professional 5.8%
Etc. 3.9% latent construct, and an AVE of 0.50 or above is deemed
M-banking used CCB 28.6% acceptable (Fornell and Larcker, 1987). As shown in
ICBC 23.7% Table 2, the factor loading value of each item is higher
CMB 20.3% than 0.63 (0.632*0.849) and significant at the 0.001
BOC 8.5% level, the CR values of constructs range from 0.786 to
ABC 7.1% 0.836 and the AVE values range from 0.540 to 0.641,
Etc. 11.8% indicating a satisfactory convergent validity.
Comfortable with Extremely uncomfortable 0.0% Discriminant validity is the extent to which the mea-
m-banking Uncomfortable 2.5% sure is not a reflection of other variables (Lee et al.,
Somewhat uncomfortable 3.2% 2007). We compared the shared variances between fac-
Neither comfortable nor 20.5%
tors with the AVE of the individual factors to assess the
uncomfortable
discriminant validity in this study. Table 3 gives the
Somewhat comfortable 29.5%
Comfortable 33.2% inter-construct correlations and the square roots of the
Extremely comfortable 11.1% AVE values. This table shows that the square roots of
the AVE values are higher than their shared variances,
representing good discriminant validity.
experience. Each questionnaire was scrutinized and To evaluate whether or not the measurement model
eliminated if it had too many missing values. can explain the actual observed data, our study
After a two-week survey, we obtained 434 valid assessed the measurement model in terms of eight dif-
responses, which fulfilled the requirements of sample ferent indices: the ratio between Chi-square and
2
size mentioned above. Among them, 39.6 percent were degrees of freedom (w /df): standard root mean-
male and 60.4 percent were female. A majority of them square residual (SRMR): root mean square error of
(60.6 percent) were in their 20s and 30s (24.4 percent approximation (RMSEA): goodness of fit index
20s; 36.2 percent 30s). In total, 75.4 percent of the (GFI): adjusted goodness of fit index (AGFI): normed
respondents were employed. The m-banking services fit index (NFI): comparative fit index (CFI): and
of CCB, ICBC, and CMB were three of the most fre- Tucker-Lewis index (TLI). As shown in Table 4, all
quently used services (28.6 percent CCB; 23.7 percent actual fit indices are within the recommended level.
ICBC; 20.3 percent CMB). Most of the respondents were The measurement model is appropriate.
comfortable with m-banking. More details about the Two tests were also conducted to examine the Com-
demographics of the sample are presented in Table 1. mon Method Variance (CMV). First, we conducted a
Harmans single-factor test (Podsakoff et al., 2003). The
Results results show that the largest variance explained by indi-
vidual factor is 34.79 percent (< 50 percent). Therefore,
The research model was tested using AMOS 20.0. Fol- none of the factors can explain the majority of the var-
lowing the two-step approach recommended by Ander- iance. Secondly, we modeled all items as the indicators
son and Gerbing (1988): we conducted Confirmatory of a factor representing the method effect and re-
Factor Analysis (CFA) to test the measurement model estimated the model (Malhotra et al., 2006). The results
and an assessment of the structural model. The maximum indicate a poor fitness, for example, the RMSEA is
likelihood approach was chosen as the model estimation 0.122 (>0.08): SRMR is 0.0948 (>0.05): and NFI is
method.
Yuan et al: An investigation of users continuance intention towards mobile banking in China 27

Table 2. Descriptive statistics and convergent validity.

Constructs Items Mean Standard deviation Factor loading Cronbachs alpha CR AVE

Perceived task-technology fit TTF1 4.735 1.568 0.767 0.828 0.832 0.623
TTF2 4.426 1.666 0.844
TTF3 4.652 1.444 0.754
Perceived usefulness PU1 5.316 1.398 0.783 0.825 0.826 0.544
PU2 5.217 1.376 0.683
PU3 5.293 1.366 0.759
PU4 5.276 1.322 0.721
Perceived ease of use PEOU1 4.350 1.348 0.763 0.836 0.842 0.641
PEOU2 4.263 1.169 0.837
PEOU3 4.247 1.080 0.800
Perceived risk PR1 3.613 1.593 0.704 0.794 0.802 0.577
PR2 3.283 1.368 0.849
PR3 2.991 1.337 0.717
Confirmation EC1 4.371 1.686 0.789 0.791 0.793 0.561
EC2 4.431 1.478 0.731
EC3 4.763 1.516 0.726
Satisfaction SAT1 4.689 1.450 0.819 0.820 0.823 0.540
SAT2 4.578 1.373 0.783
SAT3 5.101 1.219 0.632
SAT4 4.601 1.475 0.691
Continuance intention CI1 4.829 1.591 0.816 0.786 0.790 0.557
CI2 4.346 1.427 0.687
CI3 5.115 1.445 0.730

Table 3. Discriminant validity.

Perceived Perceived Perceived Perceived Continuance

task-technology fit usefulness ease of use risk Confirmation Satisfaction intention

Perceived task- 0.789

technology fit
Perceived 0.554 0.738
usefulness
Perceived ease of 0.228 0.485 0.801
use
Perceived risk 0.371 0.572 0.258 0.780
Confirmation 0.588 0.637 0.238 0.567 0.749
Satisfaction 0.481 0.615 0.200 0.547 0.674 0.735
Continuance 0.537 0.586 0.209 0.581 0.675 0.599 0.746
intention
Note: All correlations significant at p < 0.001. Diagonal elements are the square roots of AVE.

0.638 (<0.9). The results of both tests indicate that CMV GFI0.932; AGFI0.912; NFI0.921; CFI0.964;
is not a significant problem in our study. TLI0.957. These fit indices provide evidence of ade-
quate fit between the hypothesized model and the
observed data.
Structural model As shown in Table 5 and Figure 2, the explained
As Table 4 shows, the goodness-of-fit of the structural variances of perceived usefulness, satisfaction and
model is comparable to the previous CFA measurement continuance intention are 59.2 percent, 53.2 percent
model: w2/df1.758; SRMR0.041; RMSEA0.042; and 53.4 percent, respectively. The continuance
28 Information Development 32(1)

Table 4. Fit indices for measurement and structural models.

Fit indices Recommended value Measurement model Structural model

w2 343.398 372.722
df 209 212
w2/df <3 1.643 1.758
SRMR <0.05 0.037 0.041
RMSEA <0.08 0.039 0.042
GFI >0.9 0.938 0.932
AGFI >0.9 0.918 0.912
NFI >0.9 0.927 0.921
CFI >0.9 0.970 0.964
TLI >0.9 0.964 0.957

Table 5. Summary of hypotheses testing.

Standardized
Hypotheses path Coefficients T-value P-value Hypotheses testing

H1: Satisfaction!Continuance intention 0.237 3.311 <0.001 Supported

H2: Perceived usefulness!Continuance intention 0.208 2.613 0.009 Supported
H3: Perceived usefulness!Satisfaction 0.205 2.842 0.004 Supported
H4: Confirmation!Satisfaction 0.436 4.545 <0.001 Supported
H5: Confirmation!Perceived usefulness 0.483 7.048 <0.001 Supported
H6: Perceived ease of use!Continuance intention 0.070 1.252 0.210 Not supported
H7: Perceived ease of use!Perceived usefulness 0.333 6.767 <0.001 Supported
H8: Perceived task-technology fit!Continuance intention 0.225 3.601 <0.001 Supported
H9: Perceived task-technology fit!Satisfaction 0.050 0.785 0.433 Not supported
H10: Perceived task-technology fit!Perceived usefulness 0.190 3.077 0.002 Supported
H11: Perceived risk!Continuance intention 0.287 4.559 <0.001 Supported
H12: Perceived risk!Satisfaction 0.153 2.394 0.017 Supported
H13a: Satisfaction!Continuance intention moderated by gender 0.837 Not supported
H13b: Perceived usefulness!Continuance intention moderated by gender 0.274 Not supported
H13c: Perceived ease of use!Continuance intention moderated by gender N/A Not supported
H13d: Perceived task-technology fit!Continuance intention moderated by gender 0.207 Not supported
H13e: Perceived risk!Continuance intention moderated by gender 0.039 Supported

intention to use m-banking is jointly predicted by The last analysis of the data was performed to
satisfaction (b 0.237, p < 0.001), perceived use- determine the effect of gender as a moderator of con-
fulness (b 0.208, p < 0.01), perceived task- tinuance intention. The results indicate that gender
technology fit (b 0.225, p < 0.001), and perceived dose not significantly moderate the effects of satisfac-
risk (b 0.287, p < 0.001) directly, supporting H1, tion (Dw20.042, p 0.837), perceived usefulness
H2, H8, and H11, respectively. Satisfaction, in turn, is (Dw2 1.197, p 0.274), and perceived task-
predicted by perceived usefulness (b 0.205, p < 0.01), technology fit (Dw2 1.591, p 0.207) to continuance
confirmation (b 0.436, p < 0.001), and perceived risk intention, while the effect of perceived risk to
(b 0.153, p < 0.05). The results provide support for continuance intention is significantly moderated
H3, H4, and H12, respectively. In addition, confirmation by gender (Dw2 4.258, p 0.039). The detailed
(b 0.483, p < 0.001), perceived ease of use (b 0.333, statistical figures reveal that the influence of per-
p < 0.001), and perceived task-technology fit (b 0.190, ceived risk on continuance intention is stronger among
p < 0.01) have significant effects on perceived usefulness, males than females (b 0.453 and b 0.159,
supporting H5, H7, and H10, respectively. respectively).
Yuan et al: An investigation of users continuance intention towards mobile banking in China 29

Perceived task-
technology fit

0. 1
Perceived ease 0.22

90
5* **
ns.

**
of use

0.3
ns.

33*
Continuance

**
Perceived 0.208** intention
usefulness R2 = 0.534
**
* R2 = 0.592 37*
* 0.205** 0.2
8 3*
0.4 Satisfaction
*** R2 = 0.532 ***
0.436 287
0.
Confirmation

3*
.15
0
Perceived risk
Gender

Figure 2. Hypotheses testing results.

Note: * p < 0.05; ** p < 0.01; *** p < 0.001; ns. not significant.

Discussion m-banking just because it is easy to use. Since the

The results show that satisfaction is the important users become increasingly familiar with the operation
predictor of users continuance intention. The of m-banking after the initial adoption, whether the
satisfaction-intention link has previously been con- m-banking platform is easy to use or not will no
firmed in consumer behavior research over a wide longer be the focus of their attention. Even if the
range of products and service contexts (Bhattacher- providers upgrade the system, users will not be
jee, 2001a, 2001b; Chen et al., 2012; Zhou, 2013), likely to discontinue use for this reason. Therefore,
and its reconfirmation in the m-banking context fur- it is accepted that the direct effect of perceived
ther demonstrates the robustness of this association. ease of use to continuance intention is insignifi-
It means that monitoring and improving users satis- cant. However, the results in our study do not indi-
faction with m-banking is an appropriate approach cate that an easy-to-use m-banking system is
for providers to retain users. unimportant, for perceived usefulness is affected by
Confirmation is validated to have significant perceived ease of use significantly. Due to the con-
effects on perceived usefulness and satisfaction, straints of mobile terminals, users may feel it difficult
which is consistent with previous findings (e.g. Bhat- to operate m-banking (Zhou, 2012) and may spend more
tacherjee, 2001a; Lee and Kwon, 2011). Thus, banks time and effort on m-banking transactions, which will
should comprehend users expectations of m-banking, decrease their perceptions of usefulness. Therefore, it
and improve m-banking services based on users is necessary for service providers to pay more attention
requirements to better meet their expectations. Simi- to the interface and functions of m-banking. For exam-
larly, perceived usefulness is validated to be the deter- ple, banks may develop m-banking applications to
minant of satisfaction and continuance intention. replace wireless application protocol (WAP) based m-
Users are willing to use m-banking if they find it use- banking.
ful for their work, otherwise, they will switch to tradi- Although the complementarities between TTF and
tional or online banking. Hence, banks need to several models focusing on IS initial adoption (e.g.
provide new features of m-banking opportunely, pro- TAM, UTAUT, and TPB) have been confirmed in pre-
vide services accurately and with high speed, and vious studies, the appropriateness of using TTF to
ensure the stability of the system. explain continuance intention has not attracted much
Perceived ease of use, an important belief factor attention so far. In our study, the results suggest that cor-
determining IS adoption in TAM, is found to have relations between TTF and ECM constructs do exist.
no significant direct effect on continuance intention. Perceived task-technology fit strongly affects perceived
This result reflects that users will not continually use usefulness and continuance intention. If users perceive
30 Information Development 32(1)

m-banking services are unfit for their tasks, they will Theoretical and practical implications
perceive these services to be of low usefulness and form To our knowledge, this study is one of the very few that
low intention to continued usage. Thus, banks may has attempted to investigate factors affecting users con-
adhere to user-centered principles when providing tinuance intention towards m-banking. The findings of
m-banking services, to better meet the needs of users. the study have various implications for research. First,
For instance, travelers and businessmen who are always the extant studies on m-banking users continued usage
on a trip may pay close attention to reliability, security focus on their technology perceptions and rarely take the
and convenience of m-banking, while younger users task-technology fit into account. By incorporating the
may be more concerned with functions and cost. Banks TAM and TTF into the ECM, our study investigates
may subdivide users and provide different m-banking the users continuance intention from the perspectives
services to meet different subgroups needs, achieving of users perceptions towards technology and task-
a good task-technology fit. Unexpectedly, the direct technology fit. Secondly, the results which confirm that
effect of perceived task-technology fit on satisfaction perceived task-technology fit not only affects continu-
is insignificant. The possible reason may be that satis- ance intention but also affects perceived usefulness
faction comes from actual use. reveal the appropriateness and importance of using TTF
Inconsistent with the previous studies focusing on to explain continuance intention. Thirdly, by their incor-
the continued usage of m-banking, the results of our poration in the context of m-banking, our study has
study show that perceived risk has strongly negative broadened the boundaries of the ECM, TAM, and TTF,
effects on satisfaction and continuance intention. The and has enriched the IS continued usage literature. The
possible reason is that mobile commerce in China has whole model provides high explanatory power for
been developing rapidly but is still in its early stages, users continuance intention. Future research may con-
the level of mobile commerce security needs to be sider examining users continued usage of other mobile
improved, and the m-banking security issue is still the services (e.g. mobile SNS; mobile funding) based on our
focus of users attention. If users feel unsafe in con- research model.
ducting m-banking transactions, their satisfaction This study can also articulate implications for practice.
with and continuance intention of m-banking may First, the results show that perceived risk is the strongest
be reduced. Thus, m-banking service providers need predictor of users continuance intention, followed by
to improve information security technologies, provide satisfaction, perceived task-technology fit, and perceived
statements of guarantees, help desk, and a certificate, usefulness. Thus, guaranteeing the safety of m-banking
etc., so as to decrease users perceptions of risk services is crucial for banks to improve users satisfac-
towards m-banking. In addition, appropriate market- tion, and thus form their intention to continue usage. Sec-
ing is essential. Banks may demonstrate the safety ondly, we find that perceived task-technology fit has
of their m-banking services via a variety of media significant effects on perceived usefulness and continu-
(e. g. print, TV, online) and provide guidance for ance intention. This indicates that m-banking service pro-
users on how to conduct m-banking transactions prop- viders need to segregate the market and provide
erly and safely. personalized services to different user groups. Thirdly,
Moreover, our study empirically observes the mod- the significant link between perceived ease of use and
erating effects of gender. Unexpectedly, the effect of perceived usefulness suggests that it is necessary for
perceived risk is more salient for men. A plausible banks to present a user-friendly and easy-to-use m-
reason for this result may be that men are more banking system. Fourthly, the results show that men are
rational than women. When they feel it is dangerous more concerned with perceived risk than women. Conse-
to conduct transactions, it may be easier for men to quently, banks should adopt different strategies for male
no longer use the m-banking services. However, no and female customers to enhance their confidences that
matter whether the users are male or female, satisfac- using mobile banking is safe and protected.
tion is the most important determinant of their con-
tinuance intention, and perceptions of usefulness
and task-technology fit are also the focus of their con- Conclusions
cerns. Thus the results showing that gender does not Motivated by the need to better understand the motiva-
significantly moderate the effects of satisfaction, per- tions and barriers of users continuance intention
ceived usefulness and perceived task-technology fit to towards m-banking, our study builds upon the ECM,
continuance intention, seem to be acceptable. TAM, and TTF, and postulates satisfaction, perceived
Yuan et al: An investigation of users continuance intention towards mobile banking in China 31

usefulness, perceived ease of use, perceived task- PU3: M-banking lets me manage my personal
technology fit, and perceived risk as key beliefs. Satis- finances more quickly.
faction, in turn, is explained by confirmation, perceived PU4: Overall, m-banking is useful in managing my
usefulness, and perceived risk. In this study, the mea- personal finances.
surement model is confirmed with adequate conver-
gent and discriminant validity, the structural model
Perceived ease of use
provides a good fit, and most path coefficients in the
model are found statistically significant. (adapted from Davis (1989): Lin (2011)).
Our study suffers some limitations as follows: first,
PEOU1: Learning to operate m-banking is easy for
we conducted this research in the eastern provinces of
me.
China. It is conceivable that users from other regions
PEOU2: It is easy to use m-banking to accomplish
or countries may have different attitudes, perceptions,
banking transactions.
and reactions to m-banking services from those of the
PEOU3: Interaction with m-banking does not
users included in this study. We must be cautious
require a lot of mental effort.
when generalizing the findings to users beyond east-
ern China. Future research across different regions
or countries will be encouraged. Secondly, the theore- Perceived risk
tical model accounts for 53.4 percent of the variance (adapted from Kang et al. (2012)).
in continuance intention, indicating that some impor-
tant predictors may be missing. Future research may PR1: I am confidently aware of the risks associated
explore the possible effects of other factors (e.g. with m-banking.
familiarity, intimacy) to users continuance intention PR2: M-banking is dangerous for me to use.
of m-banking services. Thirdly, the research was con- PR3: There is a considerable risk involved in parti-
ducted using a short-term snapshot of users behavior. cipating in m-banking rather than other modes of
Actually, users behavior is dynamic. Future research banking services (e.g. traditional banking, online
can adopt a longitudinal study to examine users beha- banking)
viors in different time periods and make comparisons,
thus giving a clearer picture of users continued
usage. Confirmation
(adapted from Bhattacherjee (2001a, 2001b)).
Appendix A. Scales and items ECT1: My experience with using m-banking was
Perceived task-technology fit better than what I expected.
ECT2: The functions provided by m-baking were
(adapted from Goodhue and Thompson (1995): Zhou
more than what I expected.
et al. (2010)).
ECT3: Overall, most of my expectations from
TTF1: The functions of m-banking are enough to using m-banking were confirmed.
help manage my personal finances.
TTF2: The functions of m-banking are appropriate Satisfaction
to help manage my personal finances.
TTF3: In general, the functions of m-banking fully (adapted from Bhattacherjee (2001a, 2001b)).
meet my needs of finance management. How do you feel about your overall experience of
m-banking use:
Perceived usefulness SAT1: Very dissatisfied/Very satisfied.
(adapted from Bhattacherjee (2001a, 2001b): Zhou SAT2: Very displeased/Very pleased.
et al. (2010)). SAT3: Very frustrated/Very contented.
SAT4: Absolutely terrible/Absolutely delighted.
PU1: M-banking improves efficiency in managing
my personal finances.
PU2: M-banking improves convenience in manag- Continuance intention
ing my personal finances. (adapted from Bhattacherjee (2001a, 2001b)).
32 Information Development 32(1)

CI1: I intend to continue using m-banking rather Chen S.-C (2012) To use or not to use: Understanding the
than discontinue its use. factors affecting continuance intention of mobile bank-
CI2: My intention is to continue using m-banking ing. International Journal of Mobile Communications
than use any alternative means (traditional banking 10(5) 490507.
or online banking). ChinaIRN(2013) The number of mobile banking users
exceeded 340 million. http://www.chinairn.com/news/
CI3: If I could, I would like to discontinue my use
20131007/111004164.html. (Accessed 11th October
of m-banking (reverse coded). 2013).
Choi H, Kim Y and Kim J (2011) Driving factors of post
adoption behavior in mobile data services. Journal of
References Business Research 64(11): 12121217.
Anderson JC and Gerbing DW (1988) Structural equation Cruz P, Neto LBF, Munoz-Gallego P and Laukkanen T
modeling in practice: A review and recommended (2010) Mobile banking rollout in emerging markets:
two-step approach. Psychological Bulletin 103(3): Evidence from Brazil. International Journal of Bank
411423. Marketing 28(5): 342371.
AI-Jabri IM and Sohail MS (2012) Mobile banking adop- Dahlberg T, Mallat N, Ondrus J and Zmijewska A (2008)
tion: Application of diffusion of innovation theory. Past, present and future of mobile payments research:
Journal of Electronic Commerce Research 13(4): A literature review. Electronic Commerce Research and
373385. Applications 7(2): 165181.
Amin H, Hamid MRA, Lada S and Anis Z (2008) The DAmbra J, Wilson CS and Akter S (2013) Application of
adoption of mobile banking in Malaysia: The case of the task-technology fit model to structure and evaluate
Bank Islam Malaysia Berhad. International Journal of the adoption of E-books by Academics. Journal of the
Business and Society 19(2): 4353. American Society for Information Science and Technol-
Bhattacherjee A (2001a) Understanding information sys- ogy 64(1): 4864.
tems continuance: An expectation-confirmation model. Dasgupta S, Paul R and Fuloria S (2011) Factors affecting
MIS Quarterly 25(3): 351370. behavioral intentions towards mobile banking usage:
Bhattacherjee A (2001b) An empirical analysis of the ante- Empirical evidence from India. Romanian Journal of
cedents of electronic commerce service continuance. Marketing 3(1): 628.
Decision Support Systems 32(2): 201214. Davis FD (1989) Perceived usefulness, perceived ease of
Bentler P and Chou C (1987) Practical issues in structural use, and user acceptance of Information technology.
modeling. Sociological Methods & Research 16(1): MIS Quarterly 13(3): 319340.
78117. Davis FD, Bagozzi RP and Warshaw PR (1989) User
Brown I, Zaheeda C, Douglas D and Stroebel S (2003) Cell acceptance of computer technology: A comparison of
phone banking: Predictors of adoption in South Africa two theoretical models. Management Science 35(8):
An exploratory study. International Journal of Informa- 9821003.
tion Management 23(5): 381394. Dishaw M and Strong DM (1999) Extending the technol-
CEBNET(2013) The mobile banking users of three banks ogy acceptance model with task-technology fit con-
exceeded 180 million. http://www.cebnet.com.cn/2013/ structs. Information & Management 36(1): 921.
0329/124678.shtml. (Accessed 11th October 2013). Fornell C and Larcker D (1987) A second generation of
Chan S.-C and Lu M.-T (2004) Understanding Internet multivariate analysis: Classification of methods and
banking adoption and use behavior: A Hong Kong per- implications for marketing research. In Houston M. J.
spective. Journal of Global Information Management (Ed.) Review of marketing (pp. 407450). Chicago:
12(3): 2143. American Marketing Association.
Chang C.-C, Liang CY, Yan C.-F and Tseng J.-S (2013) Garbarino E and Strahilevitz M (2004) Gender differences
The impact of college students intrinsic and extrinsic in the perceived risk of buying online and the effects of
motivation on continuance intention to use English receiving a site recommendation. Journal of Business
mobile learning systems. The Asia-Pacific Education Research 57(7): 768775.
Researcher 22(2): 181192. Gefen D (2003) TAM or just plain habit: A look at experi-
Chen CS (2013) Perceived risk, usage frequency of mobile enced online shoppers. Journal of End User Computing
banking services. Managing Service Quality 23(5): 15(3): 113.
410436. Goodhue DL and Thompson RL (1995) Task-technology
Chen H and Liu LL (2012) Empirical analysis on continu- fit and individual performance. MIS Quarterly 19(2):
ous improvement of mobile content services effects on 213236.
continuance use intention. China Communications Hong S.-J, Thong JYL and Tam KY (2006) Understanding
9(12): 164172. continued information technology usage behavior: A
Yuan et al: An investigation of users continuance intention towards mobile banking in China 33

comparison of three models in the context of mobile post-adoption beliefs of mobile Internet users. Interna-
Internet. Decision Support Systems 42(3): 18191834. tional Journal of Electronic Commerce 11(4): 1151.
Hsiao WH and Chang TS (2013) Understanding consu- Lee M.-C (2010) Explaining and predicting users continu-
mers continuance intention towards mobile advertising: ance intention toward e-learning: An extension of the
A theoretical framework and empirical study. Behaviour expectation-confirmation model. Computer & Educa-
& Information Technology. DOI: 10.1080/0144929X. tion 54(2): 506516.
2013.789081. Lee MKO, Cheung CMK and Chen ZH (2007) Understand-
Junglas I, Abraham C and Watson RT (2008) Task- ing user acceptance of multimedia messaging services:
technology fit for mobile locatable information systems. An empirical study. Journal of the American Society for
Decision Support Systems 45(4): 10461057. Information and Technology 58(13): 20662077.
Kang H, Lee MJ and Lee JK (2012) Are you still with us? A Lee S, Shin B and Lee HG (2009) Understanding post-
study of the post-adoption determinants of sustained use adoption usage of mobile data services: The role of
of mobile-banking services. Journal of Organizational supplier-side variables. Journal of the Association for
Computing and Electronic Commerce 22(2): 132159. Information Systems 10(12): 860888.
Kim B (2010) An empirical investigation of mobile data Lee Y and Kwon O (2011) Intimacy, familiarity and continu-
service continuance: Incorporating the theory of planned ance intention: An extended expectation-confirmation
behavior into the expectation-confirmation model. model in web-based services. Electronic Commerce
Expert Systems with Applications 37(10): 70337039. Research and Applications 10(3): 342357.
Kim H, Lee I and Kim J (2008) Maintaining continuers vs. Limayem M and Cheung CMK (2008) Understanding
converting discontinuers: Relative importance of post- information systems continuance: The case of Internet-
adoption factors for mobile data services. International based learning technologies. Information and Manage-
Journal of Mobile Communications 6(1): 108132. ment 45(4): 227232.
Kim SS and Malhotra NK (2005) A longitudinal model of Lin H.-F (2011) An empirical investigation of mobile bank-
continued IS use: An integrative view of four mechan- ing adoption: The effect of innovation attributes and
isms underlying postadoption phenomena. Management knowledge-based trust. International Journal of Infor-
Science 51(5): 741755. mation Management 31(3): 252260.
Koenig-Lewis N, Palmer A and Moll A (2010) Predicting Lin TC, Wu S, Hsu JSC and Chou YC (2012) The integration
young consumers take up of mobile banking services. of value-based adoption and expectation-confirmation
International Journal of Banking Marketing 28(5): models: An example of IPTV continuance intention. Deci-
410432. sion Support Systems 54(1): 6375.
Koo C and Wati Y (2010) Toward an understanding of the Lin W.-S (2012) Perceived fit and satisfaction on web
mediating role of trust in mobile banking service: An learning performance: IS continuance intention and
empirical test of Indonesia case. Journal of Universal task-technology fit perspectives. International Journal
Computer Science 16(13): 18011824. of Human-Computer Studies 70(7): 498507.
Laforet S and Li X (2005) Consumers attitudes towards Luarn P and Lin HH (2005) Toward an understanding of
online and mobile banking in China. International Jour- the behavioral intention to use mobile banking. Comput-
nal of Bank Marketing 23(5): 362380. ers in Human Behavior 21(6): 873891.
Lam SY, Shankar V and Erramilli MK (2004) Customer Malhotra NK, Kim SS and Oatil A (2006) Common method
value, satisfaction, loyalty, and switching costs: An illus- variance in IS research: A comparison of alternative
tration from a business-to-business service context. Jour- approaches and a reanalysis of past research. Manage-
nal of the Academy of Marketing Science 32(3): 293311. ment Science 52(12): 18651883.
Larsen TJ, Sreb AM and Sreb (2009) The role of Moon J.-W and Kim Y.-G (2001) Extending the TAM for a
task-technology fit as users motivation to continue World-Wide-Web context. Information and Manage-
information system use. Computers in Human Behavior ment 38(4): 217230.
25(3): 778784. Morris A (2013) Report: Mobile banking users to exceed
Laukkanen T (2007) Internet vs mobile banking: Compar- 1B in 2017. http://www.fiercewireless.com/europe/
ing customer value perceptions. Business Process Man- story/report-mobile-banking-users-exceed-1b-2017/
agement Journal 13(6): 788797. 2013-01-09 (Accessed 11th October 2013).
Lee C, Cheng HK and Cheng H.-H (2007) An empirical Ng EH and Kwahk KY (2010) Examining the determinants
study of mobile commerce in insurance industry: of mobile Internet service continuance: A customer rela-
Task-technology fit and individual differences. Decision tionship development perspective. International Journal
Support Systems 43(1): 95110. of Mobile Communications 8(2): 210229.
Lee I, Choi B, Kim J and Hong SJ (2007) Culture- Nysveen H, Pedersen PE and Thorbjernsen H (2005)
technology fit: Effects of cultural characteristics on the Explaining intention to use mobile chat services:
34 Information Development 32(1)

Moderating effects of gender. Journal of Consumer Yu C.-S (2012) Factors affecting individuals to adopt
Marketing 22(4): 247256. mobile banking: Empirical evidence from the UTAUT
Park SY, Nam M.-W and Cha S.-B (2012) University stu- model. Journal of Electronic Commerce Research
dents behavioral intention to use mobile learning: Eval- 13(2): 104121.
uating the technology acceptance model. British Journal Zhao L, Lu Y, Zhang L and Chau PYK (2012) Assessing
of Educational Technology 43(4): 592605. the effects of service quality and justice on customer
Podsakoff PM, MacKenzie SB, Lee J.-Y and Podsakoff satisfaction and the continuance intention of mobile
NP (2003) Common method biases in behavioral value-added services: An empirical test of a multidi-
research: A critical review of the literature and rec- mensional model. Decision Support Systems 52(3):
ommended remedies. Journal of Applied Psychology 645656.
88(5): 879903. Zhou T (2011a) An empirical examination of initial trust in
Puschel J, Mazzon JA and Hernandez JMC (2010) Mobile mobile banking. Internet Research 21(5): 527540.
banking: Proposition of an integrated adoption intention Zhou T (2011b) Understanding mobile Internet continu-
framework. International Journal of Bank Marketing ance usage from the perspectives of UTAUT and flow.
28(5): 389409. Information Development 27(3): 207218.
Riquelme HE and Rios RE (2010) The moderating effect of Zhou T (2012) Examining mobile banking user adoption
gender in the adoption of mobile banking. International from the perspectives of trust and flow experience. Infor-
Journal of Bank Marketing 28(5): 328341. mation Technology and Management 13(1): 2737.
Shin D.-H (2007) User acceptance of mobile Internet: Zhou T (2013) An empirical examination of continuance
Implication for convergence technologies. Interacting intention of mobile payment services. Decision Support
with Computers 19(4): 472483. Systems 54(2): 10851091.
Shin YM, Lee SC, Shin B and Lee HG (2010) Examining Zhou T, Lu Y and Wang B (2010) Integrating TTF and
influencing factors of post-adoption usage of mobile Inter- UTAUT to explain mobile banking user adoption. Com-
net: Focus on the user perception of supplier-side attri- puters in Human Behavior 26(4): 760767.
butes. Information Systems Frontiers 12(5): 595606.
Sripalawat J, Thongmak M and Ngramyarn A (2011)
About the authors
M-banking in metropolitan Bangkok and a comparison
with other countries. Journal of Computer Information Shunbo Yuan is a lecturer at the School of Business, Jiax-
Systems 51(3): 6776. ing University. His research interests include mobile com-
Stevens J (2002) Applied multivariate statistics for the merce and information technology adoption. Dr. Yuan
social sciences (5th Ed.). London, Psychology Press. received his PhD in Management Science with a concen-
Suoranta M and Mattila M (2004) Mobile banking and
tration in users information behavior from the School of
consumer behavior: New insights into the diffusion
Information Management at the Nanjing University.
pattern. Journal of Financial Services Marketing
8(4): 354366. Contact: School of Business, Jiaxing University, Jiaxing,
Tan KS, Chong SC, Loh PL and Lin BS (2010) An evalua- P.R. China, 314001. Phone: 86 15888317156. Email:
tion of e-banking and m-banking adoption factors and alexshun915@gmail.com
preference in Malaysia: A case study. International
Journal of Mobile Communications 8(5): 507527. Yong Liu is an associate professor at the School of
Thong JYL, Hong SJ and Tam KY (2006) The effects of Business, Jiaxing University. Contact: School of Business,
post-adoption beliefs on the expectation-confirmation Jiaxing University, Jiaxing, P.R. China, 314001. Phone:
model for information technology continuance. Interna- 86 13325731702. Email: Sirliu2005@126.com
tional Journal of Human-Computer Studies 64(9):
799810. Ruihong Yao is a lecturer at the School of Business,
Vatanasombut B, Igbaria M, Stylianou AC and Rodgers W Jiaxing University. Contact: School of Business, Jiaxing
(2008) Information systems continuance intention of University, Jiaxing, P.R. China, 314001. Phone: 86
web-based applications customers: The case of online
15824313520. Email: yaya-mlv@163.com
banking. Information & Management 45(7): 419428.
Wu J.-H and Wang S.-C (2005) What drives mobile com-
Jing Liu is an associate professor at the School of Manage-
merce? An empirical evaluation of the revised technol-
ogy acceptance model. Information & Management ment in the Nanjing University of Posts and Telecommunica-
42(5): 719729. tions. She earned her PhD from Nanjing University. Contact:
Yang AS (2009) Exploring adoption difficulties in mobile School of Management, Nanjing University of Posts and Tel-
banking services. Canadian Journal of Administrative ecommunications, Nanjing, P.R. China, 210023. Phone: 86
Sciences 26(2): 136149. 13851503143. Email: liuj_m@njupt.edu.cn