Expert Systems With Applications 59 (2016) 33–46

Contents lists available at ScienceDirect

Expert Systems With Applications

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

Mobile technology acceptance model: An investigation using mobile

users to explore smartphone credit card
Keng-Boon Ooi a,∗, Garry Wei-Han Tan b
a
Faculty of Business & Information Science, UCSI University, Kuala Lumpur, Malaysia
b
Faculty of Business and Finance, Universiti Tunku Abdul Rahman, Kampar, Perak, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Smartphone credit card (SCC) is an emerging payment method using NFC-enabled smartphones. The
Received 30 January 2016 proximity payment allows consumers to pay their products and services by waving their smartphones
Revised 14 April 2016
with a NFC reader. While there are advantageous adopting SCC, the adoption rate has not been encour-
Accepted 15 April 2016
aging. Interestingly, existing research work on past information technology and system models have so
Available online 18 April 2016
far focused primarily on organizational context and adopted specifically for work. Furthermore, past an-
Keywords: tecedents were mainly constructed using electronic commerce literatures which do not reflect the ac-
Mobile commerce tual mobile environment. In contrast SCC is mainly adopted voluntarily by mobile users and for personal
Near field communication (NFC) purposes. Thus this leads to the difficulty in drawing meaningful conclusion. The study addresses these
Mobile technology acceptance model limitations by proposing a new mobile technology acceptance model (MTAM) which consists of mobile
(MTAM) usefulness (MU) and mobile ease of use (MEU) to determine SCC adoption. In anticipating on the com-
Mobile payment
plexity which exists in the mobile environment, additional mobile constructs namely mobile perceived
Partial least squares-structural equation
modelling-artificial neural network
security risk (MPSR), mobile perceived trust (MPT), mobile perceived compatibility (MPC) and mobile per-
(PLS-SEM-ANN) ceived financial resources (MPFR) were incorporated into the parsimonious MTAM. The integrated model
was applied to 459 mobile users through a questionnaire approach and tested using partial least square-
structural equation modelling-artificial neural network (PLS-SEM-ANN) has provided a new impact and
a possible new research methodology paradigm as it is able to capture both linear and non-linear rela-
tionships. While the model confirms the role of MU in MTAM, MEU needs for more attention in practice.
The results from the extended model showed that only three of the proposed hypotheses were non-
significant in this study and thus warrant further investigation. The study contributes to academia by
proposing new mobile constructs that brings together MTAM to assess the likelihood of mobile users to
adopt SCC. The study also offers several important managerial implications which can be generalized to
the mobile studies of other transportation, hotel, banking, and tourism industries.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction tured Supplementary Service Data, Wireless Application Protocol,

Gone were those days where traditionally payment is con-
fined to either physical exchange of notes, coins, cheques, cards
or the transfer of payment over the internet (Balachandran & Tan,
2015). The phenomenal growth in mobile technologies has cre-
ated new payment opportunities for consumers and businesses.
Payment today has now evolved to mobile devices (m-devices)
known as mobile payment (m-payment). Tode (2016) explains that
the m-payment method can be categorized into three perspectives
namely mobile remote commerce, mobile proximity payments and
mobile peer-to peer. Additionally, the technology includes Unstruc-
∗
Corresponding author. Tel.: +60391018880.
E-mail addresses: ooikengboon@gmail.com (K.-B. Ooi),
garrytanweihan@gmail.com (G.W.-H. Tan).
General Packet Radio Service and short messaging services (Tan,
Ooi, Chong, & Hew, 2014a). Each of the m-payment technology or
method however has several limitations which make payment dur-
ing checkout a hassle. The ergonomics and usability of the devices
such as small screen to display information and limited keyboard
makes m-devices awkward and unsuitable for payment adoption
(Ondrus & Pigneur, 2007). The internet connection which enables
the m-payment solutions to work may not be 100 percent sta-
ble (Valcourt, Robert, & Beaulieu, 2005). The payment solution is
also complicated due to the multiple steps that are required dur-
ing payment process (Tan et al., 2014a). Another drawback is re-
membering the pin correctly during authentication (Cheong, Ling,
& Teh, 2014). As such, consumers require more effort and time in
learning to use the applications which makes the popularity unde-
sirable. A truly mobile solution should be able to link the internet

http://dx.doi.org/10.1016/j.eswa.2016.04.015
0957-4174/© 2016 Elsevier Ltd. All rights reserved.
34 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
world and the real world of physical objects (Teh, Ahmed, Cheong,
& Yap, 2014). The emergence of smartphone credit card (SCC) over-
comes all of the limitation of traditional m-payment solutions.
SCC provides simple communication setup, ease of use, mobiquity
and requires only little power consumption (Egger, 2013; Madl-
mayr, Langer, Kantner, & Scharinger, 2008; Teh et al., 2014). SCC
can be defined as a payment of goods and services that is con-
ducted using smartphone with the aid of Near Field Communi-
cation (NFC) technology. In broad terminology, NFC works using
short range wireless technology between two devices of 10 cm or
less. By waving the NFC-enabled smartphone over a sales termi-
nal, checkout process can be completed faster using minimum ef-
fort. This is because signature is not required during the payment
process (Zmijewska, 2005; Leong, Hew, Tan, & Ooi, 2013). The cur-
rent SCC technology includes Apple Pay, Google Wallet, Visa Pay-
Wave, Simply Tapp and BellID. With benefits that will overcome
the limitation of traditional m-payment solutions, SCC is foreseen
as a vital catalyst to the existing m-payment solutions and will
potentially alter consumers’ lifestyles and change the m-payment
industry as a whole. SCC is even considered as a fourth wave of
income for mobile operators after data, voice and text plan. While
there are tremendous benefits adopting SCC as an alternative m-
payment solution, the adoption rate is far from full utilization. Teo,
Tan, Ooi, Hew, and Yew (2015a) explained that SCC has yet to
reach mass market which leads to frustration among m-payments
providers. Feijóo, Gómez-Barroso, and Ramos (2016) explain that
market analysts are still sceptical on the adoption. This is despite
the availability of SCC infrastructures since 2010 (Tan et al., 2014a).
Leong et al. (2013) elaborated on the variety of merchants adopt-
ing to SCC ranging from hypermarket, retailer, specialty store, con-
venience store, petrol station, cinema to public transportation.
As the success of SCC is dependent on the consumers’ up-
take, without a critical mass of consumers, merchants will be dis-
couraged from investing on the infrastructure. According to Teo
et al. (2015a), by 2017, global m-payment transaction is expected
to reach $1.3 trillion. Feijóo et al. (2016) further added that m-
payments are expected to increase twofold from 2012 to 2016 with
NFC technologies playing an increasing role. The introduction of
Apple Pay in 2014 which runs on NFC technology on iPhone is also
expected to represent huge business opportunities for merchants
and retailers in the coming years (Apple, 2015). Idate even esti-
mated that the volume of NFC transactions will reach €53.8 billion
Euros in 2018 from €4.6 billion in 2014 (Idate.org, 2014). In ad-
dition, NFC-enabled phone penetration is expected to reach 1907
million by 2018. With such huge potential, businesses are increas-
ing investments on the NFC infrastructures. Therefore, there is a
need to understand critical prerequisites that affect consumers’ in-
tention to adopt SCC. SCC is in essence an information technol-
ogy (IT)/information system (IS). Hence, to understand the accep-
tance process of SCC there is a need to comprehend IT/IS adoption
models. Among prominent and common IT/IS models adopted to
understand the adoption and utilization of new technological in-
novations includes “Technology Acceptance Model (TAM)”, “Diffu-
sion of Innovation (DOI)”, “Unified Theory of Acceptance and Use
of Technology (UTAUT)” and “Model of Adoption of Technology in
Households (MATH)”. The models however are not without limi-
tations. TAM asserted that the actual use is related to individual’s
attitude towards using technology and the model identified two
salient determinants of system use behaviour namely perceived
usefulness (PU) and perceived ease of use (PEOU) (Davis, 1989).
Originally, TAM was constructed to explain individual’s adoption of
electronic mail system in an organizational settings. The adoption
was based on work purposes with the individual employee of the
organization. As such the cost is normally borne by organization
with the adoption mandatory. In contrast, SCC is adopted volun-
tarily and mainly for personal purpose by individuals in a non-
organizational settings. The cost is mainly absorbed by individuals
and therefore the adoption decision is not solely based for utilitar-
ian reasons. Moreover, Tsai and LaRose (2015) stressed that users
must have the sufficient knowledge and skills to use new innova-
tion when compared to employees in organization where training
and technical support are provided. While TAM has been widely
validated in many mobile studies and able to explain many adop-
tion studies, they do not draw meaningful conclusion in view of
the differences. UTAUT was advocated by Venkatesh, Morris, Davis,
and Davis (2003) and put forward four core constructs that are di-
rect determinants intention and subsequently usage behaviour. The
constructs namely performance expectancy (PE), effort expectancy
(EE), social influence and facilitating conditions (FC) are also mod-
erated by gender, age and experience in the acceptance of IT.
Venkatesh, Thong, and Xu (2012) however criticized that the model
is again based on organizational contexts and developed primarily
based on the acceptance of technology among employees. Technol-
ogy adoption outside the workplace differs on a number of dimen-
sions such as differences in types of tasks and complexity of inter-
actions (Brown, Venkatesh, & Bala, 2006). Thus UTAUT has limita-
tion in describing mobile studies adoption such as SCC because po-
tential adopters are actual mobile consumers and not simply tech-
nology users with decisions mainly centralized from the personal
context. In view that many past IT/IS models was adopted based
on the workplace environment, Brown and Venkatesh (2005) pro-
posed Model of Adoption of Technology in Households (MATH).
The model which is based on Theory of Planned Behaviour (Ajzen,
1991) framework, focused on the adoption of technology in the
household using three constructs namely attitudinal beliefs, nor-
mative beliefs and control beliefs. While the model can explain 50
percent of the variance, the model however is limited to the ap-
plicability of adoption of personal computers (PC) for home use
(Brown et al., 2006). Teo, Tan, Cheah, Ooi, and Yew (2012) argued
that there are differences in the adoption of new technology be-
tween desktop and mobile. Finally, Roger (1983) introduces DOI
which consists of observability, relative advantage (RA), trialabil-
ity, complexity (CO) and compatibility. The model explains that
prior to the adoption of the new innovation; consumers will go
through a four stage process which includes knowledge, persua-
sion, decision and confirmation. Balachandran and Tan (2015) high-
lighted that trialability and observability in their original form are
not useful when predicting a new IT/IS adoption. Lancelot Miltgen,
Popovič, and Oliveira (2013) explained that since the new technol-
ogy has yet to reach mass wide, the majority of consumers would
not be able to test the technology and thus both models are not
well-suited. As most of the conventional frameworks has also lim-
ited ability to explain the adoption of new technologies, many re-
searchers have extended and modified the model with new con-
structs (e.g., Sim, Tan, Wong, Ooi, & Hew, 2014; Tan, Sim, Ooi, &
Kongkiti, 2012; Teo, Tan, Ooi, & Lin, 2015b). The approach helps
to increase the variance explained and making the argument for
model sufficiency. Although the constructs are rich in studies on
the factors of technology adoption, the innovation being studied is
mainly borrowed from electronic commerce literatures. A majority
of consumers however behave and react differently in an electronic
environment as compared to mobile. Mobile users perception of
ease of use behaviour for example could be heavily influenced by
the access speed, screen size, text input facilities, storage and bat-
tery life span as compared to desktop computers (Tan, Ooi, Leong,
& Lin, 2014b). Hence again they are not suited to be adopted in
mobile studies.
In order to close the research gap, the study proposed a
new base model termed as ’mobile technology acceptance model’
(MTAM) which consists of mobile usefulness (MU) and mobile ease
of use (MEU) based on the most relevant mobile technology liter-
atures. Both constructs were given priority based on the similarity
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
Fig. 1. Conceptual research framework.
of PU, PE, RA and PEOU, EE and CO in TAM,UTAUT and DOI respec-
tively, acknowledging their importance in the study of IT/IS adop-
tions. In today’s context however both constructs cannot be solely
considered as the main purpose that contributes to the customer’s
adoption. In a mobile environment, consumers’ trust, character-
istics and financial resources are equally important in determin-
ing the success rate (Leong et al., 2013; Tan et al., 2014a). Hence
the study also proposed four new mobile constructs namely mo-
bile perceived security risk (MPSR), mobile perceived compatibil-
ity (MPC), mobile perceived trust (MPT), and mobile perceived fi-
nancial resource (MPFR) to increase the overall predictability. The
combined framework represents a novel approach in comprehend-
ing SCC from a perspective of a consumer and not just from the
view point of technology. In this research, the study also adopts
behavioural intention (IU) as a proxy for consumer acceptance
(Sim et al., 2014). Sheppard, Hartwick, and Warshaw (1988) con-
clude that IU is an appropriate predictor of later usage among con-
sumers. The choice is appropriate in view that SCC is still at the
beginning phase of the product life cycle which has yet to receive
the necessary attention. Based on the new integrated framework
that brings together MTAM (see Fig. 1), the study helps to pro-
vide precise insights on SCC acceptance among consumers. The re-
search is outlines as follows. In the subsequent section, we pro-
vide an overview of the technology behind NFC and elaborate on
the past research conducted on SCC. Following this, we describe
our hypotheses developments and present our research framework.
Then, the research method and data analysis are elucidated. This
is follow by the discussion of the research findings, theoretical
and managerial implications. Finally, the paper concludes with av-
enues for future research and the limitation of study. The paper
has two main contributions. First, we developed MTAM based on
the characteristics of mobile users and specifically tailored to mo-
bile environment which so far has yet to be fully modelled in ex-
isting IT/IS research. The present model is anticipated to serve as
a main reference for future studies in mobile technologies. Sec-
ond, to date, there’s rather limited attention paid in mobile specific
antecedents. Many academicians have attempted to borrow con-
structs from electronic commerce literatures for adoption in mobile
studies. Our proposed mobile constructs overcome the shortcom-
ing and will provide a better holistic picture from the mobile adop-
35
tion perspective. The findings of the study offer important man-
agerial implications. By investigating the underlying reasons why
consumers are hesitant to adopt SCC, marketers can better formu-
late their marketing strategies to convert non-adopters to adopters.
This leads to better guidance in designing a successful implemen-
tation of SCC. Thus the new innovation can be effectively marketed
and thereby leads to a higher acceptance and effective use in fu-
ture.
2. Literature reviews
2.1. The technology behind near field communication (NFC)
NFC device operates at a frequency band of 13.56 MHz which
enables the devices to exchange data at 106, 212 or 424 Kbit/s
(Azhari, 2014). The development of NFC standard was first pro-
posed by Nokia, Sony and Phillips in 2004. Subsequently in 2006,
the first NFC-compatible m-device (Nokia 6131) was introduced
based on the specifications standards set during the 2004 NFC fo-
rum (Nearfieldcommunication.org, 2015). The commercial interests
in NFC m-devices however only become popular in 2010 when
Samsung launched Samsung Nexus S smartphone (Cheong et al.,
2014). Azhari (2014) elaborated that the technology ideally can
transmit over a distance of 20 cm but practically, the transmis-
sion only ranges between 45 cm maximally. In view of the short
distance requirements, NFC is often referred to as ’mere touch’,
’proximity wave’, ‘tap’ methods of transfer and become a popu-
lar method in exchanging quick data between devices, readers and
other NFC tags (Hoy, 2013; Volpentesta, 2015). In view of the fast,
easy and cheap implementations, commercial interests in SCC have
become popular with acceptance in different marketplace world-
wide. Telecommunications organizations for example have been
cooperating with local banks to offer NFC credit card payment
(Chen and Chang, 2013; Leong et al., 2013). Thus they are now
available for adoption in mass public transportation in Hong Kong,
Taiwan and United Kingdom (Chen & Chang, 2013; Garner, Ed-
wards, & Coulton, 2006), hotels in Sweden and Spain (Cheong et
al., 2014) and banks in London, Australia and Hong Kong (Oliveira,
Faria, Thomas, & Popovič, 2014).There are also emerging research
to explore SCC in airport (Kalakou, Psaraki-Kalouptsidi, & Moura,
2015), tourism (Egger, 2013; Pesonen & Horster, 2012) and even
home care (Cheong et al., 2014; Teh et al., 2014).
2.2. Previous research on smartphone credit card (SCC)
Reviewing past literatures, interestingly there appears to be
only a few articles that have published empirical works on
the desires of consumers to accept SCC. Balachandran and Tan
(2015) adopted 487 valid questionnaires based on a modified DOI
theory to study on the NFC mobile payment. The study integrates
amount of information, financial resources and variety of services
with relative advantage, complexity and compatibility from DOI.
Using a strict criterion to select their target respondents and a
multiple regression technique they found that only relative ad-
vantage is non-significant. Tan et al. (2014a) on the other hand
present a comprehensive model by combining technology accep-
tance model, psychological science constructs and financial related
risk to study on the intention to adopt NFC mobile credit card.
Their empirical study which includes 156 respondents from various
age groups and backgrounds were tested using structural equation
modelling and multi group analysis. Results confirmed on the ap-
plicability of the combined model and yielded that finance-related
risk is non-significant in their study. In a survey based study by
another m-payment researcher, Leong et al. (2013) combines trust
based construct, behavioural control construct, psychological sci-
ence constructs with usefulness and ease of use to study on the
36 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
intention to adopt NFC-enabled mobile credit card. Their findings
which adopted 262 consumers using neural network approach re-
mains mix which warrant further investigations. In another study,
Teo et al. (2015a) employed convenience sampling on 194 con-
sumers to gauge on their opinion on the effects of perceived trans-
action convenience (PTC) and perceived transaction speed (PTS) to
adopt m-payment. Based on the combination model of UTAUT, they
discovered that EE, FC and PTS influences adoption rate. All the
above studies provide early possibilities on the barriers in accept-
ing SCC solutions.
3. Hypotheses formulation
3.1. Mobile perceived compatibility (MPC)
MPC refers to the extent that the mobile innovation is rec-
oncilable with potential adopters past experiences, needs, value
and behavioural pattern. The adoption of SCC is not something
unusual. Consumers are already adopting to the traditional m-
payment methods using technologies such as General Packet Radio
Service, Unstructured Supplementary Service Data, Wireless Appli-
cation Protocol and short messaging services in everyday life. As
SCC is based on m-devices technology, consumers are more recep-
tive to adopt if they have experience using m-devices. Wong, Tan,
Ooi, and Lin (2015a) commented that consumers in Malaysia have
adopted smartphone to purchase products and services (Wong et
al., 2015a). In addition, consumers also using smartphone devices
to access television (Wong, Tan, Loke, & Ooi, 2014), banking (Teo
et al., 2012) and listening to music (Sim et al., 2014). NFC tech-
nology has also been adopted by consumers to exchange large
information of data, play games, share information by bumping
phones (Nearfieldcommunication.org, 2015). Balachandran and Tan
(2015) in Malaysia further explicated that consumers are adopting
m-devices to check in at locations and to share data. Thus past
experience of using smartphone devices can be regarded as an in-
dicator of mobile compatibility. Yang, Lu, Gupta, Cao, and Zhang
(2012) provided evidences that MPC influences the adoption of m-
payment in China. Similarly MPC was also found to have influence
on the adoption of mobile shopping in Malaysia whereby the im-
portance of consumers’ past belief and life style was emphasized
(Wong et al., 2015a). Based on a research conducted by Schierz,
Schilke, and Wirtz (2010), MPC was also found in influencing use-
fulness of m-payment system. Finally, as SCC runs on ’Touch ID’
and proximity payment (Apple, 2015) consumers might consider
the concept as new and hence experience a struggle to adopt. Thus
if SCC is designed to be compatible with the adopters’ past expe-
riences, this would lead to the perception that SCC is easy to use.
In view of these findings, the following hypotheses have been con-
structed:
H1: MPC has a positive and significant relationship with MU.
H7: MPC has a positive and significant relationship with IU.
H9: MPC has a positive and significant relationship with MEU.
3.2. Mobile perceived financial resources (MPFR)
MPFR refers to the extend individuals belies that the he or she
has the financial resources to use SCC (e.g., cost of device, transac-
tion fees and bank specific charges). Wong et al. (2014) commented
that financial resources usually hamper the adoption of new in-
novation among consumers. SCC devices are normally pricier than
traditional phones. This is due to the higher cost in production
which requires a redesigning of traditional models and therefore
expensive to produce. Some of the m-devices could come with
bundle contract such as 24-months contract (Tan et al., 2014a).
Consumers would also need to pay for transactional costs for usage
(BusinessNewsDaily.com, 2015). The adoption of SCC can also come
with annual credit card fees and charges (Hsbc.com.hk, 2015).
Given the high financial considerations to adopt SCC, the technol-
ogy may be out of reach for consumers with limited and constraint
financial resources. The practical benefits and easiness of use on
the adoption would therefore diminish. Based on 262 valid ques-
tionnaires, Leong et al. (2013) provide evidences that MPFR is not
a significant factor with MU and MEOU in their study on NFC-
enabled mobile credit card acceptance in Malaysia. Finally accord-
ing to Balachandran and Tan (2015), the adoption of SCC includes
conversion, maintenance and accessing costs. They found that the
costs prevent consumers from the IU to adopt NFC mobile payment
in Malaysia and suggested to reduce the cost through creative pro-
motion and pricing techniques. Thus the following hypotheses are
being constructed:
H2: MPFR has a negative and significant relationship with MEU.
H8: MPFR has a negative and significant relationship with IU.
H10: MPFR has a negative and significant relationship with MU.
3.3. Mobile ease of use (MEU)
MEU which has the same meaning as ease of use and complex-
ity of using a particular IT/IS (Venkatesh et al., 2003). Therefore
MEU combines the constructs mentioned above and refers to the
perception of complexity to learn and use for potential adopters
when adopting to m-devices. The adoption of SCC requires min-
imum effort to use and learn when compared to the old fash-
ion m-payment methods. Thus this leads to favourable impressions
for potential adopters. By selecting Apple Pay and placing the in-
dividual’s finger on the ‘Touch ID’, payment could be conducted
in a breeze (Apple, 2015). Consumers also do not need to require
much effort in registering SCC prior to usage. Consumers can regis-
ter their credit card details such as name card number and expiry
date using the camera icon in their smartphone (Appleinsider.com,
2014). When consumers feel that SCC is easy to use and learn; they
will more likely to adopt this new innovation. Dutot (2015) ex-
plained that when consumers start to adopt a new system, they
tend to stress on the possibility of learning success and in this
case directly affect their IU to use a particular technology. MEU
was found to influence the IU to adopt contactless m-payment and
m-payment in Malaysia (Teo et al., 2015a, 2015b). In another study
in France using 320 smartphone users based on PLS-SEM analysis
on NFC mobile payment, MEU was found to be significant with IU
(Dutot, 2015). These studies stressed on the friendly interface, sys-
tem operation and function in ensuring higher users’ acceptance.
Hence the next hypothesis has been constructed:
H3: MEU has a positive and significant relationship with IU.
3.4. Mobile usefulness (MU)
MU is similar with sub-constructs of usefulness such as rel-
ative advantage and extrinsic motivation and explains how they
lead to improvement in job performance for the particular individ-
ual (Venkatesh et al., 2003). Likewise in this research, MU refers
to the perception of usefulness enhancement rendered for poten-
tial adopters when adopting to m-devices. Compared with tradi-
tional m-payment methods, SCC which is based on contactless pay-
ment technology results to effectiveness in payment such as con-
venient and quicker checkout. Itbusinessedge.com (2015) explained
that payment can be made by placing the individual’s iPhone near
the NFC reader. Thus no card swiping is involved. In addition, mul-
tiple credit cards can be loaded in SCC. This allows consumers the
flexibility to choose their credit card for each specific transaction
(Nearfieldcommunication.org, 2015). The ability to store multiple
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
cards in SCC reduces the risk of losing credit cards. A study by
Tan et al. (2014a) in Malaysia on NFC m-devices found that MU
leads to higher intention to adopt for m-payment. Similar result
was also illustrated by Teo et al. (2015b) whereby the adoption
of m-payment brings time saving and portability. Interestingly MU
was found to be non-significant according to the study by Dutot
(2015) and Balachandran and Tan (2015) in France and Malaysia
respectively despite the advantages rendered by NFC m-payment.
In view of these contradicting findings, the following hypothesis
has been constructed:
H4: MU has a positive and significant relationship with IU.
3.5. Mobile perceived security risk (MPSR)
MPSR is a vital construct in the research of wireless environ-
ment as it determine the decision of consumers whether to adopt
a new innovation. MPSR refers to the perception of protection
against risk associated with mobile transactions. This could be risk
of losing confidential data which leads to financial losses. Mail
Online (2015), reported on a new breed of digital pick pocket-
ing known as digital skimming. By using an app and standing 6
inches away, data can be stolen during any NFC transactions within
the store. Thus, the NFC infrastructure needs to be configured cor-
rectly by merchants and retailers to avoid any misuse or attacks
(Hoy, 2013). Zupanovic (2015) stressed that NFC technology per-
mits consumers to experience safer interactions due to its bet-
ter encryption of transacted data. In another study on NFC smart-
phone access control system based on 40 participants, Cheong et
al. (2014) found that MPSR has significant relationship with IU.
While MPSR is considered to be important, interestingly findings
by Tan et al. (2014a) found that perceived risk does not influence
the IU of NFC-enabled m-devices in Malaysia. Thus, the construct
requires further investigation. The following hypothesis has been
constructed:
H5: MPSR has a negative and significant relationship with IU.
3.6. Mobile perceived trust (MPT)
MPT refers to a psychological expectation that others will be
sincere in keeping promises and will not behave opportunistically
in expectation of a promised service derived from mobile transac-
tions. If the retailers show benevolence, integrity and competence,
consumers are more likely to engage with SCC transactions (Leong
et al., 2013). According to Yang and Lin (2014), it is more diffi-
cult for a retailer and consumer to build online trust compared
to an offline trust. This is caused by the absent of any face-to-
face interactions between both parties during any online transac-
tions. Hence MPT is an important factor in the online environment
such as in any mobile transactions. Otherwise relationships cannot
be established over the long-term due the uncertainty. Addition-
ally online environment promotes betrayal, lack of accountability
and opportunity for deceit due to its association with risk char-
acteristics (Immonen & Sintonen, 2015). Sztompka (1999) pointed
out that trustworthiness is based on the trustee’s reputation and
performance. Tamimi and Sebastianelli (2015) remarks that privacy
seals, third party assurances, security policies, provide feedbacks
help to increase the consumer’s perception on online trustwor-
thiness. Qualitative interviews by Mallat (2007) provide evidences
that trust in reliable providers reduced risks in m-payment ser-
vices. Based on past researchers, MPT has been proven to increase
intention for online payments in China (Yang, Pang, Liu, Yen, &
Michael Tarn, 2015). Similar Teo et al. (2015b) also found support-
ing evidences on the influential effect of MPT and IU. They con-
cluded that consumer would only transact if the IT/IS is free of
37
identity theft and mobile fraud. The following hypothesis has been
constructed:
H6: MPT has a positive and significant relationship with IU.
4. Research method
4.1. Sampling procedure and data collection
A paper-based survey was conducted between October and
November 2015 at five major shopping malls at Klang Valley,
Malaysia to validate the research model. Klang Valley was cho-
sen since area has high population density with about 1/5 of
Malaysian’s total population (Pan, Chew, Cheah, Wong, & Tan,
2015). The sampling location is also a major hub for economic,
financial and culture in the county with a fair balance of ma-
jor ethnics in Malaysia such as Malay, Chinese and Indians
(Wordpopulationreview.com, 2016). Thus this would give a fair
characteristics presentational of the gender, age and ethnicity of
the general population in Malaysia which are relevant to this con-
text of study. The five major shopping malls were chosen due to
the high number of potential mobile users available which will ex-
pedite our process in data collection. The questionnaire distribu-
tion was assisted by four research assistants and was based on
systematic sampling method whereby, every second participants
who set foot in the shopping malls were approached and inquired
to voluntarily complete the questionnaires. Prior to distribution
the participants were brief on the purpose of study and guaran-
teed anonymity. In addition, to avoid a respond bias in the sur-
vey, names were excluded. Upon agreement, a short video clip
on smartphone credit card (e.g., https://www.youtube.com/watch?
v=4I9MbIrlEUw) was presented. The approach has been widely
adopted in recent years by international researchers (Tan et al.,
2014b). The samples were subsequently briefed on the terminol-
ogy and definitions adopted in the questionnaires. While SCC in-
cludes different technologies and can be applied in various sectors
we explained that SCC refers to the overall concepts and was mea-
sured from a general perspective. The respondents were also en-
couraged to ask the research assistants on the questionnaire items
that they didn’t understand prior to answering the questionnaire.
This would help to eliminate the possibility of respondents’ filling
out the questions without exactly understanding the content of the
questionnaire. On a whole, 501 questionnaires were handed out.
The questionnaires were scrutinized and only 459 valid and usable
samples were adopted, yielding a valid response rate of 91.61 per-
cent among volunteered participants.
4.2. Instrument development
The items in the survey instrument were based on our review
of existing and past literatures relevant to the research model.
Some of the wordings were modified and reworded to fit the con-
text of SCC. The research were anchored on a seven-point Likert-
type scale measurement varying from “1 (strongly disagree)” to “7
(strongly agree)”. A pre-test was also performed with three mobile
commerce scholars with a background on mobile payment to en-
sure that the questionnaires have no semantic problems. Some of
the clarity of questions, length of instruments, the completeness
of content and structure which were found to be ambiguous and
vogue were further amended based on the provided feedback. The
instruments were then further pilot-tested with 20 mobile pay-
ment users. The participant group has knowledge on the mobile
payment systems in Malaysia. Slight modifications were made to
the wording resulted from the pre-test. The multiple phases of in-
strument development helped to establish the appropriate level
38 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46

Table 1
Questionnaire source and number of items.

Constructs Number of items Sources

Mobile Usefulness (MU)
Mobile Ease of Use (MEU)
Mobile Perceived Compatibility (MPC)
Mobile Perceived Trust (MPT)
Mobile Perceived Security Risk (MPSR)
Mobile Perceived Financial Resources (MPFR)
Intention to use (IU)
5 Davis et al. (1989) and Tan et al. (2014a)
5 Davis et al. (1989) and Tan et al. (2014a)
5 Rogers (1983) and Wong et al. (2015a)
5 Zhou and Lu (2011) and Leong et al. (2013)
3 Tan et al. (2014a)
4 Leong et al. (2013)
5 Tan et al. (2014a)

of internal and content validity. The survey questionnaire com- Table 2

Demographic profile of respondents.
posed of two sub-sections. The first focused exclusively on the re-
spondents’ demographic characteristics. The second focused on the Item Frequency Percent
main constructs of the model with 31 questions asked. Items for
Gender Male 198 43.1
measuring MU and MEU were modified using a scale derived from Female 261 56.9
the original TAM study by Davis, Bagozzi, and Warshaw (1989) and Age Below 20 32 7.0
Tan et al. (2014a) because of the sustainability in this context of 21–25 111 24.2
study. The measurement of MPC were taken from those developed 26–30 105 22.9
31–35 88 19.2
and validated by Wong et al. (2015a) and the original DOI model
36–40 52 11.3
(Rogers, 1983). Items for measuring MPT were adapted from Zhou Above 40 71 15.5
and Lu (2011) and Leong et al. (2013). MPSR was measured using Marital status Single 289 63.0
items originated from Tan et al. (2014a). With regards to MPFR, Married 170 37.0
the construct were measured using items derived from Leong et Highest education No college degree 122 26.6
level
al. (2013). Finally measurements for IU were taken from Tan et al.
Diploma/Advanced diploma 136 29.6
(2014a). Table 1 shows the final measurement instrument used and Bachelor degree/Professional 138 30.1
their sources of operationalization. qualification
Postgraduate qualification 12 2.6
Others 51 11.1
5. Data analysis Respondent’s Banking 54 11.8
industry
5.1. Profile of respondents Financial institutions 44 9.6
IT-related 28 6.1
Manufacturing 21 4.6
The data presented in Table 2 provides the demographic details Retail 46 10.0
on the gender, age, marital status, highest level of academic qualifi- Telecommunications 13 2.8
cation of the sample respondents. Of our samples, 198 respondents Tourism 7 1.5
were male and 261 female which indicates that the number of fe- Education 35 7.6
Others 211 46.0
males is larger than the number of males. Majority of the respon-
Period of credit Less than 3 years 245 53.4
dents are relatively young with 54.1% below the age of 31. Only card use
15.5% were older than 40 years old. In view of the m-payment 3–6 years 120 26.1
nature, this unbalance is understandable, because the majority of Over 6 years 94 20.5
the mobile users are usually well educated, young at age and are Frequency of credit 1–3 times 137 29.8
card use (per
able to afford a computer (Shin, 2010). The results also show that month)
a large number of the respondents have at least a basic college 4–10 times 155 33.8
education or above at 62.3 percent. Only 26.6 percent of respon- 11–20 times 30 6.5
dents have no formal education. Regarding the respondent’s indus- Others 137 29.9
try, the largest proportion of the respondents belongs to the oth-
ers industry, followed by banking, retail, financial institutions, edu-
cation, IT, manufacturing, telecommunication and tourism. Notice- for studies with small sample size, and does not require normal-
ably, 46.6 percent of respondents have a period of credit card us- ity to be demonstrated (Aibinu & Al-Lawati, 2010). Secondly PLS is
age of more than 3 years whereas the majority is less than 3 years. suited when analyzing predictive models with multifaceted con-
In terms of the frequency of credit card usage, 63.6 percent of the structs (Ang, Ramayah, & Amin, 2015). Thirdly PLS can simulta-
respondents reported a usage of less than 11 times in a month. neously analyze the structural and measurement model (Hulland,
The characteristics of the demographic profiles are important in- 1999). Thus, PLS can offer higher degree of statistical power com-
dicators that would help retailers and merchants to communicate pared to the SEM (Hair, Hult, Ringle, & Sarstedt, 2014). Therefore,
easily with their target respondents. we have decided to use PLS instead of the SEM for the prelim-
inary analysis of the two-stage PLS-ANN approach. Furthermore,
5.2. Statistical analyses the two-stage PLS-ANN approach was engaged based on two rea-
sons. First, PLS is a variance-based SEM analysis that is incapable
The research model displayed in Fig. 1 was assessed by using of detecting non-linear relationships (Leong et al., 2013). Based on
Partial Least Square-Structural Equation Modelling-Artificial Neural the ANOVA test of deviation from linearity (Hew, Lee, Ooi, & Wei,
Network (PLS-SEM-ANN) approach (Leong, Hew, Lee, & Ooi, 2015). 2015; Tan et al., 2014b), we found that there were non-linear re-
The significant determinants identified from the PLS analysis were lationships between MU, MPSR and MPT with IU. This weakness
used as the input neuron for the Artificial Neural Network (ANN) is complemented by the ANN analysis which can detect both lin-
analysis. There are three reasons why PLS is preferred in compar- ear and non-linear relationships (Teo et al., 2015a). Secondly, even
ison to SEM for the PLS-ANN approach. Firstly PLS is appropriate though ANN has the ability to recognize both linear and non-linear
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46 39

relationships (Leong et al., 2015), however due to the ‘black box’ Table 3
Item loadings, AVE, CR, and Cronbach’s α .
operation of the ANN algorithm, it is not suitable for parametric
hypothesis testing (Sim et al., 2014). Hence, in order to take ad- Constructs AVE CR Cronbach’s α Item loadings
vantage of the both, we have used a combined PLS-ANN approach
IU 0.8787 0.9731 0.9654 IU1 0.9435
as this approach will enable us to test the hypothesis and on the IU2 0.9536
same time detect both linear and non-linear relationships that ex- IU3 0.9460
ist in the research model. IU4 0.9058
Outer measurement model was firstly assessed for both reliabil- IU5 0.9372
MEU 0.7503 0.9375 0.9166 MEU1 0.8227
ity and validity; while inner structural model was later on exam-
MEU2 0.8794
ined. This is known as two-step approach, and it is consistent with MEU3 0.8957
Wong, Tan, Hew, and Ooi (2016). Moreover, mediating test was MEU4 0.8482
also conducted and reported afterward. There are seven latent vari- MEU5 0.8829
ables and 32 observed variables in the proposed model. Through MPC 0.8190 0.9577 0.9447 MPC1 0.8870
MPC2 0.9208
the calculator by Soper (2015), which is suggested by Westland
MPC3 0.9015
(2010,2012), the recommended sample size is 88. The following MPC4 0.9231
settings were made in the calculator: MPC5 0.8920
MPFR 0.7594 0.9264 0.9031 MPFR1 0.8829
• anticipated effect size - 0.3 (Dutot, 2015) MPFR2 0.9444
MPFR3 0.8464
• desired statistical power level - 0.8 (Cohen, 1988)
MPFR4 0.8062
• significance probability level - 0.05 (Fisher, 1925) MPSR 0.9104 0.9682 0.9511 MPSR1 0.9424
MPSR2 0.9586
In addition, Wong (2013) posited that under the most extreme MPSR3 0.9614
situation, wherein there are 10 arrows pointing to a construct, 91 MPT 0.7786 0.9462 0.9293 MPT1 0.8860
samples would be required. In view of these, the sample size of MPT2 0.8944
MPT3 0.8517
this study, 459, is considered sufficient. MPT4 0.8910
MPT5 0.8882
MU 0.7894 0.9493 0.9332 MU1 0.8829
5.2.1. Assessing outer measurement model MU2 0.8937
Reliability and validity of the outer measurement model were MU3 0.8576
assessed by item reliability, convergent validity, and discriminant MU4 0.9142
MU5 0.8932
validity. Following the measures suggested by both Kim and Park
(2013), and Chen and Shen (2015), Cronbach’s α and composite re-
liability (CR) are the measures to ascertain reliability; average vari-
Table 4
ance extracted (AVE) and item loadings are used to evaluate con-
Square root of the AVE and construct correlations.
vergent validity; while square root of the AVE was assessed to eval-
uate discriminant validity. Fransen et al. (2014), p.1376) referred IU MEU MPC MPFR MPSR MPT MU

convergent and discriminant validity as “the extent to which differ- IU 0.9374

ent measures of the same construct are in fact related” and “how MEU 0.5299 0.8662
different measures of different constructs are not related” respec- MPC 0.5950 0.6472 0.9050
tively. Cronbach’s α should exceed the threshold of 0.70, as sug- MPFR 0.0504 0.0878 0.0484 0.8714
MPSR 0.2410 0.1924 0.3014 0.0672 0.9541
gested by Nunnally and Bernstein (1994), while CR should achieve MPT 0.5938 0.4904 0.5515 0.0279 0.3228 0.8824
0.60 and more, in accordance with Bagozzi and Yi (1988). The MU 0.5462 0.7363 0.6287 0.1281 0.2115 0.5038 0.8885
readings tabled in Table 3 demonstrate that reliability has been
Notes: Off-diagonal elements are the construct correlations, while diagonal elements
achieved for the measurement model. Moreover, Kline (1988) ad- in bold and italic are the square root of the AVE.
vised that AVE should not be lesser than 0.50, while Fornell and
Larcker (1981) opined that item loading has to be 0.70 and above.
It is then concluded that convergent validity is shown in the mea- Table 5
Hypotheses testing results.
surement model. Lastly, Kim and Park (2013) believed that discrim-
inant validity is acquired if the square root of the AVE exceeded Hypotheses Paths Path coefficients T Statistics Remarks
the correlations between the constructs. As illustrated in Table 4,
H1 MPC -> IU 0.2595∗∗∗ 3.4611 Supported
all the square root of the AVE are greater than the construct corre- H2 MPFR -> MEU 0.0566 1.2268 Not supported
lations, which implies that discriminant validity has been acquired H3 MEU -> IU 0.0882 0.9903 Not supported
too. H4 MU -> IU 0.1505∗ 1.9607 Supported
H5 MPSR -> IU 0.0077 0.1701 Not supported
H6 MPT -> IU 0.3290∗∗∗ 5.6895 Supported
5.2.2. Examining inner structural model H7 MPC -> MU 0.6240∗∗∗ 14.9058 Supported
H8 MPFR -> IU 0.0011 0.0251 Not supported
The outcomes of the hypotheses testing are listed in Table 5 and
H9 MPC -> MEU 0.6444∗∗∗ 15.6089 Supported
illustrated in Fig. 2 too. There are only six hypotheses supported, H10 MPFR -> MU 0.0979∗ 2.1359 Supported
namely H1 (β = 0.2595, p < 0.001), H4 (β = 0.1505, p < 0.05), H6 ∗∗∗
Notes: p < 0.001; ∗ p < 0.05; R2 = 0.4806 (IU); R2 = 0.4220(MEU); R2 = 0.4049 (MU).
(β = 0.3290, p < 0.001), H7 (β = 0.6240, p < 0.001), H9 (β = 0.6444,
p < 0.001), and H10 (β = 0.0979, p < 0.05). All influences exerted by
MPC on MU (R2 = 0.4049), MEU (R2 = 0.4220), and IU (R2 = 0.4806)
are significant; while MPFR has impact on MU only. Other than 5.2.3. Evaluating the mediation effects
MPC; MU and MPT serve as the significant predictors of IU. It As endorsed by Zhang, Benyoucef, and Zhao (2016), this study
should be noted that MPC has larger effect on MU than its coun- employs the method proposed by Baron and Kenny (1986) in eval-
terpart, and serves as the sole predictor for MEU. In addition, MPT uating the mediating roles of MU and MEU. Three regressions
has the strongest influence on IU compared to MPC and MU. equations need to be estimated separately under this method:
40 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46

Table 6
Mediation test and Sobel’s test.

IV MV DV IV -> DV (a) IV -> MV (b) IV + MV -> DV (c) Mediating effect Sobel’s z

IV -> DV MV -> DV

MPC MU IU 0.5954∗∗∗ 0.6304∗∗∗ 0.4163∗∗∗ 0.2854∗∗∗ Partial 4.0495∗∗∗

MPFR MU IU 0.1030NS 0.1263∗ 0.0245NS 0.5427∗∗∗ No N/A
MPC MEU IU 0.5954∗∗∗ 0.6471∗∗∗ 0.4339∗∗∗ 0.2493∗∗∗ Partial 3.3039∗∗∗
MPFR MEU IU 0.1030NS 0.1078NS 0.0385NS 0.5250∗∗∗ No N/A
∗∗∗ ∗
Notes: p < 0.001; p < 0.05; NS = not significant; N/A = not applicable.

Table 7
RMSE values of the ANN models.

Input: MPC, MPFR Input: MPC Input: MPC, MPT, MU

Output: MU Output: MEU Output: IU

Network Training Testing Training Testing Training Testing

1 0.1435 0.1465 0.1162 0.1141 0.1354 0.1093

2 0.1502 0.1282 0.1171 0.1142 0.1340 0.1113
3 0.1425 0.1205 0.1176 0.1163 0.1348 0.1178
4 0.1474 0.1243 0.1173 0.1158 0.1308 0.1127
5 0.1443 0.1221 0.1206 0.1233 0.1281 0.1042
6 0.1440 0.1161 0.1135 0.1125 0.1335 0.1102
7 0.1430 0.1200 0.1191 0.1135 0.1338 0.1116
8 0.1481 0.1201 0.1150 0.1181 0.1313 0.1053
9 0.1452 0.1206 0.1149 0.1152 0.1292 0.1123
10 0.1461 0.1151 0.1155 0.1160 0.1283 0.1075
Mean 0.1454 0.1233 0.1167 0.1159 0.1319 0.1102
Std. dev. 0.0025 0.0090 0.0021 0.0031 0.0027 0.0039

5.2.4. Artificial neural network analysis (ANN)

Fig. 2. Hypotheses testing results.
1 Regression (a) comprises of independent variable and depen-
dent variable.
2 Regression (b) should only consist of independent variable and
the mediator.
3 Regression (c) contains of the mediator, independent variable,
and dependent variable.
Following the rules set by Baron and Kenny (1986), mediation
can only be established when both Regression (a) and Regression
(b) are statistically significant, and mediator has significant influ-
ence on dependent variable under Regression (c). Partial media-
tion can be concluded, if the effect of independent variable on
dependent variable in Regression (c) is lower than Regression (a);
while full mediation exists when the linkage between independent
variable and dependent variable is significant in Regressions (a),
but turns to insignificant under Regression (c).Moreover, as rec-
ommended by Chen, Wu, Peng, and Yeh (2015), Sobel test was
also performed to validate the mediation effects played by MU and
MEU. The results for these tests are shown in Table 6. Both MU and
MEU partially mediate the effect of MPC on IU, but not the effect of
MPFR on IU. The Sobel test also delivered significant results, which
proved that the mediations to be statistically significant. This test
was only performed; if the rules set by Baron and Kenny (1986) are
fulfilled.
ANN is “a machine that is invented to model the manner in
which human brain performs a specific task or function” (Haykin,
2004, p. 24). It has the ability to examine non-compensatory de-
cision processes (Svozil, Kvasnicka, & Pospichal, 1997) and does
not require the normality assumption while providing better pre-
diction in comparison to other conventional regression techniques
(Chiang, Zhang, & Zhou, 2006). To evaluate the relative importance
of the predictors, multi-layer perceptron (MLP) was engaged in the
training and testing phases. To avoid over-fitting, a 10-fold cross-
validation procedure with 90% of data for training and the rest for
testing was deployed. Sigmoid function were used as the activa-
tion functions for the hidden and output layers while the number
of hidden layers where automatically generated by SPSS. The pre-
dictive accuracy was assessed based on the values of Root Mean
Square of Error (RMSE). Table 7 shows that the ANN models are
able to predict with a very high degree of accuracy as the RMSE
value are very small.
In order to measure the predictive power of the predictors for
IU, sensitivity analysis was engaged whereby we have computed
the normalized importance based on the percentage of the relative
importance over the largest relative importance among the predic-
tors in each ANN models. Table 8 shows that for the predictors of
MU, MPC is the strongest predictor followed by MPFR. Since the
normalized importance for MPFR is less than 10%, we may con-
clude that the effect of MPFR is very small in comparison to MU.
Since MEU has only one predictor, MPC contributed all the variance
in MEU with 100% normalized importance. Finally, the prediction
on IU is contributed mostly by MPT followed by MPC and MU with
87% and 75.3% of normalized importance respectively. The three
ANN models engaged in the analysis are shown in Fig. 3.
6. Discussion
The result of these findings indicate that MPC has a significant
relationship with MU. The study corroborates with past empirical
works conducted by Schierz et al. (2010) on m-payment system.
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46 41

Table 8
Normalized importance of predictors in ANN models.

Relative importance
Output: MU Output: MEU Output: IU

Network/Input MPC MPFR MPC MPC MPT MU

1 0.929 0.071 1.0 0 0 0.313 0.326 0.361

2 0.907 0.093 1.0 0 0 0.337 0.417 0.215
3 0.945 0.055 1.0 0 0 0.278 0.401 0.321
4 0.954 0.046 1.0 0 0 0.331 0.375 0.294
5 0.942 0.058 1.0 0 0 0.317 0.397 0.286
6 0.932 0.086 1.0 0 0 0.357 0.386 0.257
7 0.915 0.085 1.0 0 0 0.330 0.412 0.258
8 0.863 0.137 1.0 0 0 0.350 0.365 0.285
9 0.935 0.065 1.0 0 0 0.362 0.345 0.293
10 0.972 0.028 1.0 0 0 0.331 0.376 0.292
Average 0.929 0.072 1.0 0 0 0.331 0.380 0.286
Normalized importance (%) 100.0 7.8 100.0 87.0 100.0 75.3

The finding explains that when SCC fits well with the consumers’
lifestyle, they will find that SCC beneficial. For example, if con-
sumers are already adopting to the traditional payment method
such as mobile peer-to peer, a shift to a newer payment method
such as proximity payment will lead to the perception that SCC
is more efficient during payment. This is because consumers are
able to compare the usefulness rendered by both m-payment sys-
tems. The finding also confirms the study by Balachandran and
Tan (2015) whereby MPC was found to have significant influence
with IU. This could be due to the experience of consumers in us-
ing m-devices for m-commerce purposes (Wong et al., 2015a). Fur-
ther, the study also revealed that MPC has significant relationship
with MEU. Thus, it implies that consumers will find SCC easy to
use when the innovation is reconcilable with the potential users’
current habits, needs, value and past experiences.
Previous studies in Malaysia and Taiwan on NFC-enabled mobile
credit card and mobile services, indicated that MPFR has no signifi-
cant relationship with MEU (Leong et al., 2013; Wang, Lin, & Luarn,
2006). The study coincides with our findings whereby MPFR was
found to be non-significant with MEU. This means that the per-
ception of financial resources will not reduce the operational com-
plexity of SCC. On the contrary our results confirmed that MPFR
is significant with MU. The result found empirical support from
Sim et al. (2014) and Wang et al. (2006) on the study on mo-
bile technologies. This leads us to believe that when SCC is value
for money, potential adopters will have higher expectation on the
benefits of usefulness in their daily lives. Interestingly MPFR is a
non-significant determinant of IU. Our findings validate the view
by Tan et al. (2014a) on NFC mobile credit card in Malaysia. Teo
et al. (2015b) accredited the reason with the affordable prices of
smartphones and telecommunication plans in Malaysia. Similarly
Wong et al. (2015a) mentioned that the cost of m-devices has re-
duced significantly over the last few years due to the intense com-
petition among manufactures and telecommunication providers.
Regarding MEU, our study highlight that the construct is not a
significant predictor for IU. The findings contradicts with all the
past studies conducted on NFC m-devices where consumers would
accept the new innovation if the system is effortless (Balachandran
& Tan, 2015; Dutot, 2015; Leong et al., 2013; Tan et al., 2014a). The
finding is also in disagreement with past findings conducted on
mobile studies for shopping and learning (Wong, Lee, Lim, Chua, &
Tan, 2012; Wong, Tan, Loke, & Ooi, 2015c). One possible explana-
tion is that the number of steps needed to perform a SCC trans-
action is lesser than other m-payment systems. Therefore majority
of the consumers perceived that SCC is user friendly to use and
probably requires lesser effort to learn.
Fig. 3. The ANN models in the study. MU on the other hand was found to be significant with
IU. The findings coincides with the past study performed by
42 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
Tan et al. (2014a) which highlighted that the more benefits per-
ceived by mobile payment adopters, the more favourable towards
consumers’ IU. MU has also been proven to be a predictor of IU in
other mobile studies such mobile learning (Tan et al., 2012), mobile
advertising (Wong, Tan, Tan, & Ooi, 2015b), mobile banking (Teo et
al., 2012), mobile marketing (Pan et al., 2015), mobile healthcare
(Tan, Siah, Ooi, Hew, & Chong, 2015) and mobile television (Wong
et al., 2016). In this context, consumers place high importance on
time saving, convenience and safety features rendered by SCC as
contrast to offline payment.
On the contrary, MPSR is not significant with IU which the find-
ings coincides with the results conducted by Tan et al. (2014a) and
Cheong et al. (2014).One possible explanation is when compared
to other traditional mobile payment methods; SCC is portrayed
to have better encryption (Zupanovic, 2015). Thus consumers have
confidence that their personal information will not be intercepted
during any transaction and hence do not view MPSR as important.
Finally it was found that MPT is significant with IU to adopt
SCC. Similar results were found by mobile payment scholar such
as Teo et al. (2015b) whereby MPT was found to be the most influ-
ential construct on IU to adopt m-payment. This implies that con-
sumers will be more likely to adopt SCC if they believe that the
banks and retailers are honest, sincere in keeping promises and
take responsibility in the eventuality of any identity theft, fraud
or security issues.
7. Implications
7.1. Theoretical implications
The main contribution of the research paper is the development
of MTAM which comprises of MEU and MU. MTAM is more com-
prehensive model in understanding the adoption of SCC in com-
parison with traditional IT/IS models as it examines the adoption
based on actual mobile users in a non-organizational settings. Tra-
ditional IT/IS model may not be suitable as they are based on or-
ganizational settings and related to work purposes which often is
compulsory. The basic framework of MTAM would allow other con-
structs to be incorporated easily to better understand the adop-
tion of mobile technologies from the perspective of mobile con-
sumers. Further, the general applicability of the model can be ap-
plied across different mobile technologies and users. Second as
most past constructs were developed from e-commerce theories
and adopted for usage in mobile studies, the research may not
reflect the actual mobile environment. This is because SCC offers
unique characteristics which are different from the electronic com-
merce environment. As such consumers would behave differently
in view of the differences. The study contributes by introducing
four new mobile constructs namely MPSR, MPT, MPC and MPFR.
They are introduced since consumers’ choice of payment depends
on their trust, characteristics and financial resources. The new mo-
bile constructs provides a more precise explanation on the adop-
tion and investigation of SCC. Future researcher should be able to
readily relate the new constructs with other mobile technology
studies. Third, research on m-payment is mostly focused on tra-
ditional technologies. SCC is currently an emerging research topic
among academicians and scholars. We contribute to the existing
literatures on m-payment by providing information on the barriers
that prevents consumers to adopt. Lastly, one of the vital theoret-
ical contributions of this study is the use of expert systems and
artificial intelligent systems via the PLS-ANN approach in compari-
son to other related papers which used pure ANN approach as the
ANN approach has a "black box" operation that renders it not suit-
able for hypothesis testing. However with the new PLS-SEM-ANN
approach proposed in this study, we are able to address this limi-
tation.
7.2. Managerial implications
The cost of developing an infrastructure for NFC is not cheap.
Thus careful planning is essential before implementing. Based on
traditional m-payment methods, SCC offers advantages such as
quicker checkout which translate to time saving and convenience
for consumers. In addition, SCC promotes safety as consumers do
not need to carry cash when shopping. Since MU is significant with
IU, in order to improve MU, merchants and retailers should pro-
mote the advantages of SCC in their mobile marketing campaigns.
Strategic alliances should also be created between mobile compa-
nies and merchants. Apple Pay for example has teamed up with
over 220,0 0 0 stores from paying gas, amusement park, to vending
machines in the United States (Digital Trends, 2015). This creates a
favourable impression that the adoption will bring usefulness and
productivity for consumers in their job performance. Apart from
m-payment, the NFC technology can also be adopted for various
functions. According to Leong et al. (2013). p. 5605), “NFC-enabled
mobile phones may also read NFC tags that can host a diversity
of content such as vouchers, videos, social media pages, compe-
titions, travel information and etc.”. Thus, another strategy is to
integrate NFC functions in every life situations of consumers. The
strategy will lead to conversion of non-users to potential users in
the long run when they find that NFC is useful. Next when de-
veloping NFC apps, businesses need to consider consumer’s expec-
tations on the functions and usability of SCC. Businesses should
improve on functions based on suggestions from consumers such
as MEU interface so that consumers can easily operate and trans-
act with minimum effort even though they are non-significant in
this study. Apple Pay’s credit card picture in m-devices for ex-
ample can be adjusted; tap, drag and move forward or backward
to improve view (Appleinsider.com, 2014). The registration pro-
cess should also be simple without much procedure. According to
Appleinsider.com (2014), the registration of credit card using Ap-
ple Pay includes camera icon whereby consumers can scan the
name, card number, expiration date. Supports should also be ex-
tended not only through offline but also online. Automated phone
call, blogs, online FAQ, video instructions, competent personal are
some of the several aspects which can help to increase the per-
ception of EE among consumers. MPSR should also not been ne-
glected although the study indicates otherwise. When compared
with e-commerce, transactions that build upon wireless environ-
ment are more vulnerable for security attacks and interceptions.
One way to reduce the perception of MPSR is to adopt advanced
technology infrastructure. This includes adopting wireless encryp-
tion technologies, sign-on security systems and other reliable secu-
rity measures. Hsbc.com.hk (2015) for example reported that their
SCC includes pin before payment for security reasons and the con-
tactless payment features automatically turn off after 30 s when
not in use. Businesses should also adopt third-party legal certifica-
tions and continuously improvement mechanism to increase pay-
ment security. Strict security policies should also be adhered to
guarantee security during payment. Retailers should make efforts
to verify the identity of consumers prior to transaction. In addition,
safe shopping guarantee and other protection mechanism should
be properly communicated to all consumers. The guarantee and
protection mechanism should contain description that are not too
lengthy and should avoid the use of technical jargons which makes
consumers comprehension difficult. Lee, Au, and Law (2012) even
suggested communicating security policies in video format as part
of the communication with consumers. Consumers should also
be encouraged to adopt specific security measures to secure and
safeguard their private information. Based on the suggestions by
Nearfieldcommunication.org (2015), they includes only (1) con-
ducting transactions with merchants that adopts secure channels,
(2) m-devices should be in active-passive pairing and (3) adopting
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
password or any keypad locks when SCC device is turned on. When
consumers can integrate SCC in their daily activities, this will lead
to a favourable perception of MPC that they are easy to use and
has many advantages and thus influence the adoption intention.
Therefore, mobile service providers should develop and advertise
SCC in a manner that matches consumer’s prior experience, need
and behavioural pattern. Among the advantages of adopting SCC
is usefulness and convenience. Since SCC render faster checkout
during payment process (Tan et al., 2014a), the innovation should
be marketed as a lifestyle products. Thus this will suit busy con-
sumers who are always ‘time starved’ and constantly looking to
complete their tasks with minimum time. Businesses should also
focus by highlighting on the values of SCC as compared to tradi-
tional m-payment services. By analysing consumer’s values, a more
personalized marketing campaign can be emphasized. Consumers
perceived that using SCC is more expensive than traditional credit
card. This is due to the perception on of high financial cost asso-
ciated with using SCC. SCC-enabled m-devices are not cheap (e.g.,
Samsung Galaxy, iPhone). In addition to the cost of device, con-
sumers need to pay annual fees of credit card. There are also addi-
tional transaction costs involved (BusinessNewsDaily.com, 2015). In
order to ease consumers concern on MPFR, appropriate promotion
and pricing strategies should be created to reduce the financial
barrier facing potential consumers. Banks for example could lower
their credit card annual fees and commission for the first few
years. Once consumers discover the benefits and get used to the al-
ternative payment method, they will be inclined to continue usage.
Retailers are also encouraged to absorb some of the transaction
costs rather than passing the cost to consumers. Telecommuni-
cation providers could offer subsidized NFC-enabled m-devices to
spur adoption rate. Finally in order to increase MPT, businesses for
example should encouraged their clients to post up their indepen-
dent reviews on the discussion board, social network sites, blogs
and etc. since the construct is significant in this study. The commu-
nication should be properly managed with prompt responds. This
is relevant for increasing consumer’s trust as some consumer based
their decisions on the online reviews. Banks should also allow the
possibility of free trial of SCC prior to adoption. Once tried, this di-
minishes the uncertainty faced by consumers and will lead to an
increase of use. Banks, telecommunication providers are also sug-
gested to be transparent regarding the issue of privacy. Customers
need to be informed on what basis their private information is be-
ing shared. The information should be provided on the websites
for consumers’ viewing. Apple for example has uploaded their pri-
vacy overview on their website (Apple.com, 2015). All of this can
help to build trust with consumers which will lead to increase of
reputation in the long run and higher intention to adopt SCC.
8. Limitations and future directions
Our research has several weaknesses that future studies should
address. First the majority of the respondents in our demographic
information are slightly biased towards younger group with 54.1%
of them below the age of 30.Although the official data from
Malaysia Communication and Multimedia Commission (MCMC)
(2014) indicates that majority of m-devices in Malaysia are young
users, we cannot claim that it is representative and therefore can-
not be generalize to other areas or the elder. Many studies have
found that older respondents are reluctant to adopt new innova-
tion due to their risk adverse and lack of familiarity with technol-
ogy (Venkatesh et al., 2003). Additionally young consumers have
generally tended to be early adopters of new technology (Tan,
Chong, Ooi, & Chong, 2010). Future research may explore different
age groups to see if there are possible differences in IU. Second our
research model measure IU as the dependent variable rather than
actual behaviour given that the nature of SCC services which is still
43
at an infancy stage of implementation. The assumption is based
that IU is closely related to actual behaviour (Venkatesh & Davis,
20 0 0) with the method closely shared by many studies conducted
on mobile adoptions. However the actual adoption behaviour may
not necessary be reflected by the IU with Wang et al. (2006). p.
175) for example reported weak links in their study. Hence fu-
ture studies could be improved by employing actual behaviour in
the study to closely observe the later stage of SCC implementa-
tion. Third, SCC technologies range from Apple Pay, Google Wal-
let, Visa PayWave, SimplyTapp to BellID. Moreover SCC can also
be applied in various sectors such as transportation, hotel, bank-
ing, tourism and etc. The study however did not focus on any
specific SCC technology, no on any specified sector and was mea-
sured from a general perspective. Based on different scenarios and
circumstances, the perception for the same service could be dif-
ferent. While this may not necessary affect the results since the
study’s intention is to understand the motivation of consumers to
adopt SCC, the study provides some direction for future research.
Fourth, although the survey was conducted at five major shopping
malls and covers the main financial and economic hub in Malaysia,
the study may suffer from biases derived from the geographical
context of the current study. This is because the respondents of
the study are from Malaysia. While the findings is applicable to
other South East Asia countries which shares the similar character-
istics and demographics, the findings may not necessary be appli-
cable other emerging countries that lagged behind ICT infrastruc-
ture or with cross-culture (e.g., individualism, collectivism) differ-
ences. Hence future studies should conduct cross-national research
such as by concentrating on consumers from different countries to
strengthen and validate the research. Fifth, the study is restricted
to only six attributes to understand on the consumers’ intention to
adopt SCC. There are obviously other important factors that have
been excluded which might influence IU. Research for example
shown that influences system characteristics, government supports,
past adoption behaviour or other psychological behaviours influ-
ences adoption. In addition different demographic variables such as
income, education qualification or individual innovativeness may
also influence adoption (Sim, Tan, Ooi, & Lee, 2011). Future stud-
ies should therefore integrate or extend the current theoretical
model to discuss any discrepancies in terms of results rendered by
the unrevealed constructs. Finally the study focuses on SCC which
is at the early adoption stage and is an on-going process. This
cross-sectional study did not portray how consumers’ attitude can
change over time (Tan et al., 2012). Davis et al. (1989) for example
revealed that ease of use of technology adoption was significant
after fourteen weeks of implementation contradicting to their ear-
lier findings. Kim, Mirusmonov, and Lee (2010) echoed the similar
sentiments whereby they concluded that the usefulness has no ef-
fect on m-payment systems for early adopters since useful features
cannot be expected from the technology at the initial stage. Thus
future studies can employ a longitudinal study at the later stage of
SCC implementation as it would show the changes in consumer’s
behaviour.
9. Conclusion
The research conducted on mobile studies for decades have
been mainly centred on electronic commerce constructs. At the
same time findings are also inconclusive due to the inconsistency
of theoretical models applied across different mobile studies. Fur-
thermore consumers react differently in an electronic environment
as oppose to mobile. These lead to the difficulty in comparing the
research findings and drawing meaningful conclusion. Recognizing
the needs for a specific mobile constructs and model the study
draws on theories grounded from mobile technology literatures
and proposed a new MTAM framework. The proposed framework
44 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46

was integrated with additional mobile constructs such as MPC, Apple.com. (2015). Apple Pay security and privacy overview – Apple Support. Avail-
MPSR, MPT and MPFT to increase predictive power. The frame- able at: https://support.apple.com/en-us/HT203027 (Retrieved 7 March 2015) .
Appleinsider.com. (2014). How to set up Apple Pay on an iPhone 6 or 6 Plus running
work was also applied to mobile users as approach to many past iOS.1, and where you can use it right now. Available at: http://appleinsider.
models which adopted respondents from organizations. The frame- com/articles/14/10/20/how- to- set- up- apple- pay- on- an- iphone- 6- or- 6- plus-
work which is exclusively tailored from the mobile perspective running- ios- 81- and- where- you- can- use- it- right- now (Retrieved 5 March
2015).
provides insights and valuable information for businesses, govern- Azhari, F. (2014). Quick detection of NFC vulnerability. Information Management &
ment agencies, retailers, merchants, banks in designing their mo- Computer Security, 22(2), 134–140.
bile marketing strategies and the development and provisions of Bagozzi, R., & Yi, Y. (1988). On the evaluation of structural equation models. Journal
of the Academy of Marketing Science, 16(1), 74–94.
mobile services with effective technological designs.
Balachandran, D., & Tan, G. W. H. (2015). Regression modelling of predicting NFC
mobile payment adoption in Malaysia. International Journal of Modelling in Op-
erations Management, 5(2), 100–116.
Baron, R., & Kenny, D. (1986). The moderator-mediator variable distinction in so-
Acknowledgement cial psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51(6), 1173–1182.
This is a revised version of a paper presented at the 6th In- Brown, S., Venkatesh, V., & Bala, H. (2006). Household technology use: Integrating
household life cycle and the model of adoption of technology in households.
ternational Research Symposium in Service Management (IRSSM- The Information Society, 22(4), 205–218.
6), UiTM, Kuching, Malaysia, 11-15 August 2015 by Garry Wei-Han Brown, S. A., & Venkatesh, V. (2005). Model of adoption of technology in the house-
Tan and Keng-Boon Ooi and was awarded the ‘Emerald Literati hold: A baseline model test and extension incorporating household life cycle.
MIS Quarterly, 29, 399–426.
2015 Highly Commended Paper Award’ by Journal of Service Man- BusinessNewsDaily.com (2015). 27 Ways to accept mobile payments. Avail-
agement, Emerald (UK). In addition, this research is supported by able at: http://www.businessnewsdaily.com/4457- mobile- payment- solutions.
the Universiti Tunku Abdul Rahman Research Publication Scheme htm (Retrieved 6 March 2015).
Chen, J., & Shen, X. (2015). Consumers’ decisions in social commerce context: An
(UTARRPS), under the project number 6251/G03. empirical investigation. Decision Support Systems, 79, 55–64.
Chen, K., & Chang, M. (2013). User acceptance of ‘near field communication’ mobile
phone service: An investigation based on the ‘unified theory of acceptance and
use of technology’ model. The Service Industries Journal, 33(6), 609–623.
Appendix A Chen, Y., Wu, J., Peng, L., & Yeh, R. (2015). Consumer benefit creation in online group
buying: The social capital and platform synergy effect and the mediating role
of participation. Electronic Commerce Research and Applications, 14(6), 499–513.
Cheong, S., Ling, H., & Teh, P. L. (2014). Secure encrypted steganography graphical
Acronyms password scheme for near field communication smartphone access control sys-
ANN Artificial Neural Network tem. Expert Systems with Applications, 41(7), 3561–3568.
AVE Average Variance Extracted Chiang, W. Y. K., Zhang, D., & Zhou, L. (2006). Predicting and explaining patronage
CO Complexity behavior toward web and traditional stores using neural networks: A compara-
CR Composite Reliability tive analysis with logistic regression. Decision Support Systems, 41(2), 514–531.
DOI Diffusion of Innovation Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, N.J.:
L. Erlbaum Associates.
EE Effort Expectancy
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance
IS Information System
of information technology. MIS Quarterly, 13(3), 319–339.
IT Information Technology Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer
IU Behavioural Intention technology: A comparison of two theoretical models. Management Science, 35(8),
MATH Model of Adoption of Technology in Households 982–1003.
M-device Mobile Device Digital Trends (2015). Visa boosts mobile payment security in Europe, ready for
MEU Mobile Ease Of Use eventual Apple Pay launch. Available at: http://www.digitaltrends.com/mobile/
MLP Multi-layer Perceptron apple- pay- partners- news/ (Retrieved 5 March 2015).
M-payment Mobile Payment Dutot, V. (2015). Factors influencing near field communication (NFC) adoption: An
MPC Mobile Perceived Compatibility extended TAM approach. The Journal of High Technology Management Research,
MPFR Mobile Perceived Financial Resources 26(1), 45–57.
MPSR Mobile Perceived Security Risk Egger, R. (2013). The impact of near field communication on tourism. Journal of Hos-
MPT Mobile Perceived Trust pitality and Tourism Technology, 4(2), 119–133.
Feijóo, C., Gómez-Barroso, J., & Ramos, S. (2016). Techno-economic implications of
MTAM Mobile Technology Acceptance Model
the mass-market uptake of mobile data services: Requirements for next gener-
MU Mobile Usefulness
ation mobile networks. Telematics and Informatics, 33(2), 600–612.
NFC Near Field Communication
Fisher, R. (1925). Statistical methods for research workers. Edinburgh, UK: Oliver &
PE Performance Expectancy Boyd.
PEOU Perceived Ease Of Use Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unob-
PLS-SEM-ANN Partial Least Square-Structural Equation servable variables and measurement error. Journal of Marketing Research, 18(1),
Modelling-Artificial Neural Network 39–50.
PTC Perceived Transaction Convenience Fransen, J., D’Hondt, E., Bourgois, J., Vaeyens, R., Philippaerts, R. M., &
PTS Perceived Transaction Speed Lenoir, M. (2014). Motor competence assessment in children: Convergent and
PU Perceived Usefulness discriminant validity between the BOT-2 Short Form and KTK testing batteries.
RA Relative Advantage Research in Developmental Disabilities, 35(6), 1375–1383.
RMSE Root Mean Square of Error Garner, P., Edwards, R., & Coulton, P. (2006). XEPS – Enabling card-based payment
SCC Smartphone Credit Card for mobile terminals. 2006 IEEE International Symposium on Consumer Electronics.
doi:10.1109/isce.2006.1689509.
TAM Technology Acceptance Model
Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2014). A primer on partial least
UTAUT Unified Theory of Acceptance and Use of
squares structural equation modeling (PLS-SEM). Los Angeles, CA: Sage Publica-
Technology
tions.
Haykin, S. (2004). Neural networks: A comprehensive foundation (2nd ed.). Upper Sad-
References dle. River, NJ: Prentice Hall.
Hew, J. J., Lee, V. H., Ooi, K. B., & Wei, J. (2015). What catalyses mobile apps usage
Aibinu, A., & Al-Lawati, A. (2010). Using PLS-SEM technique to model construction intention: An empirical analysis. Industrial Management & Data Systems, 115(7),
organizations’ willingness to participate in e-bidding. Automation in Construc- 1269–1291.
tion, 19(6), 714–724. Hoy, M. (2013). Near field communication: Getting in touch with mobile users. Med-
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human ical Reference Services Quarterly, 32(3), 351–357.
Decision Processes, 50, 179–211. Hsbc.com.hk (2015). HSBC Mobile Payments. Available at: https://www.hsbc.com.hk/
Ang, M., Ramayah, T., & Amin, H. (2015). A theory of planned behavior perspective 1/2/hk/cards/mobilepayments (Retrieved 6 March 2015).
on hiring Malaysians with disabilities. Equality, Diversity and Inclusion: An Inter- Hulland, J. (1999). Use of partial least squares (PLS) in strategic management re-
national Journal, 34(3), 186–200. search: A review of four recent studies. Strategic Management Journal, 20(2),
Apple (2015). Apple – Apple Pay. Available at: https://www.apple.com/apple-pay/ 195–204.
(Retrieved 4 March 2015).
 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46
Idate.org (2014). Mobile and online payment. Available at: http://www.idate.org/en/
News/Mobile- and- online- payment_881.html (Retrieved 5 March 2015).
Immonen, M., & Sintonen, S. (2015). Evolution of technology perceptions over time.
Information Technology & People, 28(3), 589–606.
Itbusinessedge.com (2015). Can we trust the security of mobile pay-
ment solutions? Available at: http://www.itbusinessedge.com/articles/
can- we- trust- the- security- of- mobile- payment- solutions.html (Retrieved 7 Tan, G. W. H., Ooi, K. B., Leong, L. Y., & Lin, B. (2014). Predicting the drivers of behav-
March 2015).
Kalakou, S., Psaraki-Kalouptsidi, V., & Moura, F. (2015). Future airport terminals:
New technologies promise capacity gains. Journal of Air Transport Management,
42, 203–212.
Kim, C., Mirusmonov, M., & Lee, I. (2010). An empirical examination of factors in-
fluencing the intention to use mobile payment. Computers in Human Behavior,
26(3), 310–322.
Kim, S., & Park, H. (2013). Effects of various characteristics of social commerce (s–
commerce) on consumers’ trust and trust performance. International Journal of
Information Management, 33(2), 318–332.
Kline, R. (1988). Principles and practice of structural equation modelling (1st ed.). New
York: Guilford Press.
Lancelot Miltgen, C., Popovič, A., & Oliveira, T. (2013). Determinants of end-user ac-
ceptance of biometrics: Integrating the “Big 3” of technology acceptance with
privacy context. Decision Support Systems, 56, 103–114.
Lee, H., Au, N., & Law, R. (2012). Presentation formats of policy statements on ho-
tel websites and privacy concerns: A multimedia learning theory perspective.
Journal of Hospitality & Tourism Research, 37(4), 470–489.
Leong, L. Y., Hew, T. S., Lee, V. H., & Ooi, K. B. (2015). An SEM–artificial-neural-net-
work analysis of the relationships between SERVPERF, customer satisfaction and
loyalty among low-cost and full-service airline. Expert Systems with Applications,
42(19), 6620–6634.
Leong, L. Y., Hew, T. S., Tan, G. W. H., & Ooi, K. B. (2013). Predicting the determinants
of the NFC-enabled mobile credit card acceptance: A neural networks approach.
Expert Systems with Applications, 40(14), 5604–5620.
Madlmayr, G., Langer, J., Kantner, C., & Scharinger, J. (2008). NFC devices: Security
and privacy. In The third international conference on availability, reliability and
security (pp. 642–647).
Mail Online.. Hacker shows how easy it is to steal credit card numbers from thin air .
Malaysia Communications and Multimedia Commission (MCMC) (2014). Hand phone
users survey: Statistical brief number twelve. Available at: http://www.skmm.gov.
my/skmmgovmy/media/General/pdf/MCMC- Hand- Phone- User19112015.pdf (Re-
trieved 13 January 2016).
Mallat, N. (2007). Exploring consumer adoption of mobile payments – A qualitative
study. The Journal of Strategic Information Systems, 16(4), 413–432.
Nearfieldcommunication.org. (2015). Security Concerns with NFC Technology –
NearFieldCommunication.org. Available at: http://www.nearfieldcommunication.
org/nfc-security.html (Retrieved 6 March 2015).
Nunnally, J., & Bernstein, I. (1994). Psychometric theory (3rd ed.). New York: Mc-
Graw-Hill.
Oliveira, T., Faria, M., Thomas, M., & Popovič, A. (2014). Extending the understand-
ing of mobile banking adoption: When UTAUT meets TTF and ITM. International
Journal of Information Management, 34(5), 689–703.
Ondrus, J., & Pigneur, Y. (2007). An assessment of NFC for future mobile payment
systems. International Conference on the Management of Mobile Business (ICMB
2007). doi:10.1109/icmb.2007.9.
Pan, V. Q., Chew, P. Q., Cheah, A. S. G., Wong, C. H., & Tan, G. W. H. (2015). Mobile
marketing in the 21st century: A partial least squares structural equation mod-
elling approach. International Journal of Modelling in Operations Management,
5(2), 83–93.
Pesonen, J., & Horster, E. (2012). Near field communication technology in tourism.
Tourism Management Perspectives, 4, 11–18.
Rogers, E. (1983). Diffusion of innovations. New York: Free Press.
Schierz, P., Schilke, O., & Wirtz, B. (2010). Understanding consumer acceptance of
mobile payment services: An empirical analysis. Electronic Commerce Research
And Applications, 9(3), 209–216.
Sheppard, B. H., Hartwick, J., & Warshaw, P. R. (1988). The theory of reasoned action:
A meta-analysis of past research with recommendations for modifications and
future research. Journal of Consumer Research, 15(3), 325–343.
Shin, D. (2010). Modeling the interaction of users and mobile payment system: Con-
ceptual framework. International Journal of Human-Computer Interaction, 26(10),
917–940.
Sim, J. J., Tan, G. W. H., Ooi, K. B., & Lee, V. H. (2011). Exploring the individual char-
acteristics on the adoption of broadband: An empirical analysis. International
Journal of Network and Mobile Technologies, 2(1), 1–14.
Sim, J. J., Tan, G. W. H., Wong, J. C. J., Ooi, K. B., & Hew, T. S. (2014). Under-
standing and predicting the motivators of mobile music acceptance – A multi-
-stage MRA-artificial neural network approach. Telematics and Informatics, 31(4),
569–584.
Soper, D.. The free statistics calculators website Retrieved from (Retrieved 6 March
2015) http://www.danielsoper.com/statcalc3/calc.aspx?id=89.
Svozil, D., Kvasnicka, V., & Pospichal, J. (1997). Introduction to multi-layer feed-for-
ward neural networks. Chemometrics and Intelligent Laboratory Systems, 39(1),
43–62.
Sztompka, P. (1999). Trust: A sociological theory. Cambridge: Cambridge University
Press.
Tamimi, N., & Sebastianelli, R. (2015). The relative importance of e-tailer website at-
tributes on the likelihood of online purchase. Internet Research, 25(2), 169–183.
45
Tan, G. W. H., Chong, C. K., Ooi, K. B., & Chong, A. Y. L. (2010). The adoption of online
banking in Malaysia: An empirical analysis. International Journal of Business and
Management Science, 3(2), 169–193.
Tan, G. W. H., Ooi, K. B., Chong, S. C., & Hew, T. S. (2014). NFC mobile credit card:
The next frontier of mobile payment? Telematics and Informatics, 31(2), 292–307.
ioral intention to use mobile learning: A hybrid SEM-Neural Networks approach.
Computers in Human Behavior, 36, 198–213.
Tan, G. W. H., Siah, M. W., Ooi, K. B., Hew, T. S., & Chong, A. Y. L. (2015). The adop-
tion of PDA for future healthcare system: An emerging market perspective. In-
ternational Journal of Mobile Communications, 13(1), 1–28.
Tan, G. W. H., Sim, J. J., Ooi, K. B., & Kongkiti, P. (2012). Determinants of mobile
learning adoption: An empirical analysis. Journal of Computer Information Sys-
tem, 52(3), 82–91.
Teh, P. L., Ahmed, P. K., Cheong, S. N., & Yap, W. J. (2014). Age-group differences in
near field communication smartphone. Industrial Management & Data Systems,
114(3), 484–502.
Teo, A. C., Tan, G. W. H., Cheah, C. M., Ooi, K. B., & Yew, K. T. (2012). Can the demo-
graphic and subjective norms influence the adoption of mobile banking? Inter-
national Journal of Mobile Communications, 10(6), 578–597.
Teo, A. C., Tan, G. W. H., Ooi, K. B., Hew, T. S., & Yew, K. T. (2015). The effects of
convenience and speed in m-payment. Industrial Management & Data Systems,
115(2), 311–331.
Teo, A. C., Tan, G. W. H., Ooi, K. B., & Lin, B. (2015). Why consumers adopt mo-
bile payment? A partial least squares structural equation modelling (PLS-SEM)
approach. International Journal of Mobile Communications, 13(5), 478–497.
Tode, C.. Proximity payment is fastest growing segment of mobile payments: Forrester
- Mobile Commerce Daily – Research. Mobilecommercedaily.com .
Tsai, H., & LaRose, R. (2015). Broadband Internet adoption and utilization in the in-
ner city: A comparison of competing theories. Computers in Human Behavior, 51,
344–355.
Valcourt, E., Robert, J., & Beaulieu, F. (2005). Investigating mobile payment: Sup-
porting technologies, methods, and use. Wimob’2005), IEEE International Confer-
ence On Wireless and Mobile Computing, Networking and Communications, 2005.
doi:10.1109/wimob.2005.1512946.
Venkatesh, V., & Davis, F. (20 0 0). A theoretical extension of the technology ac-
ceptance model: Four longitudinal field studies. Management Science, 46(2),
186–204.
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: Toward a unified view. MIS Quarterly, 27(3), 425–478.
Venkatesh, V., Thong, J., & Xu, X. (2012). Consumer acceptance and use of informa-
tion technology: Extending the unified theory of acceptance and use of technol-
ogy. MIS Quarterly, 36(1), 157–178.
Volpentesta, A. (2015). A framework for human interaction with mobiquitous ser-
vices in a smart environment. Computers in Human Behavior, 50, 177–185.
Wang, Y. S., Lin, H. H., & Luarn, P. (2006). Predicting consumer intention to use
mobile services. Information Systems Journal, 16(2), 157–179.
Westland, J. C. (2010). Lower bounds on sample size in structural equation model-
ing. Electronic Commerce Research and Applications, 9(6), 476–487.
Westland, J. C. (2012). Erratum to “Lower bounds on sample size in structural equa-
tion modeling” [Electron. Commerce Res. Appl. 9 (6) (2010) 476–487]. Electronic
Commerce Research and Applications, 11(4), 445. doi:10.1016/j.elerap.2012.06.001.
Wong, C. H., Lee, H. S., Lim, Y. H., Chua, B. H., & Tan, G. W. H. (2012). Predicting the
consumers’ intention to adopt mobile-shopping: An emerging market perspec-
tive. International Journal of Network and Mobile Technologies, 3(3), 24–39.
Wong, C. H., Tan, G. W. H., Hew, T. S., & Ooi, K. B. (2016). Can mobile TV be a
new revolution in the television industry? Computers in Human Behavior, 55,
764–776.
Wong, C. H., Tan, G. W. H., Loke, S. P., & Ooi, K. B. (2014). Mobile TV: A new form
of entertainment? Industrial Management & Data Systems, 114(7), 1050–1067.
Wong, C. H., Tan, G. W. H., Ooi, K. B., & Lin, B. (2015). Mobile shopping: The next
frontier of the shopping industry? An emerging market perspective. Interna-
tional Journal of Mobile Communications, 13(1), 92–112.
Wong, C. H., Tan, G. W. H., Tan, B. I., & Ooi, K. B. (2015). Mobile advertising: The
changing landscape of the advertising industry. Telematics and Informatics, 32(4),
720–734.
Wong, C. H., Tan, G. W. H., Loke, S. P., & Ooi, K. B. (2015). Adoption of mobile social
networking sites for learning? Online Information Review, 39(6), 762–778.
Wong, K. K. (2013). Partial Least Squares Structural Equation Modeling (PLS-SEM)
Techniques Using SmartPLS. Marketing Bulletin, 24, 1–32.
Worldpopulationreview.com.. Kuala Lumpur Population 2015 – World Population Re-
view .
Yang, H., & Lin, C. (2014). Why do people stick to Facebook web site? A value theo-
ry-based view. Information Technology & People, 27(1), 21–37.
Yang, Q., Pang, C., Liu, L., Yen, D., & Michael Tarn, J. (2015). Exploring consumer
perceived risk and trust for online payments: An empirical study in China’s
younger generation. Computers in Human Behavior, 50, 9–24.
Yang, S., Lu, Y., Gupta, S., Cao, Y., & Zhang, R. (2012). Mobile payment services adop-
tion across time: An empirical study of the effects of behavioral beliefs, social
influences, and personal traits. Computers in Human Behavior, 28(1), 129–142.
Zhang, K. Z. K., Benyoucef, M., & Zhao, S. J. (2016). Building brand loyalty in so-
cial commerce: The case of brand microblogs. Electronic Commerce Research and
Applications, 15, 14–25.
46 K.-B. Ooi, G.W.-H. Tan / Expert Systems With Applications 59 (2016) 33–46

Zhou, T., & Lu, Y. (2011). The effects of personality traits on user acceptance of Zupanovic, D. (2015). Implementation model for near field communication in croa-
mobile commerce. International Journal of Human-Computer Interaction, 27(6), tian ferry ticketing system. Procedia Engineering, 100, 1396–1404.
545–561.
Zmijewska, A. (2005). Evaluating wireless technologies in mobile payments – A cus-
tomer centric approach. In Proceedings of the fourth International Conference on
Mobile Business (ICMB).