Information Systems ] (]]]]) ]]]–]]]

1 Contents lists available at ScienceDirect

3
Information Systems
5

7 journal homepage: www.elsevier.com/locate/infosys

11
Technology adoption: A conjoint analysis of consumers'
13 preference on future online banking services
15
Q1 Samson Yusuf Dauda n,1, Jongsu Lee
17 Seoul National University, Technology Management, Economics and Policy Program (TEMEP), Administrative Office (International Energy/
IT Policy Program), Bldg #37-307, 1 Gwanak-ro, Gwanak-gu, 151-742 Seoul, Republic of Korea

19

21 a r t i c l e i n f o abstract

23 Article history: The importance of service delivery technology and online service adoption and usage in
Received 3 March 2015 the banking industry has received an increased discussion in the literature in recent years.
Received in revised form Owing to the fact that Strong online banking services are important drivers for bank
25 24 April 2015
performance and customer service delivery; several studies have been carried out on
Accepted 25 April 2015
online banking service adoption or acceptance where services are already deployed and
27 Recommended by: Prof. D. Shasha
on the factors that influence customers' adoption and use or intention to use those
services. However, despite the increasing discussion in the literatures, no attempt has
29 Keywords: been made to look at consumers' preference in terms of future online banking service
Technology adoption
adoption. This study used conjoint analysis and stated preference methods with discrete
Conjoint analysis
31 choice model to analyze the technology adoption pattern regarding consumers' pre-
Stated preference
Discrete choice ference for potential future online banking services in the Nigerian banking industry. The
33 result revealed that to increase efficiency and strengthen competitiveness, banks need to
promote smart and practical branded services especially self-services at the same time
promote a universal adoption of e-banking system services that add entertainment or
35
extra convenience to customers such as ease of usage including digital wallet, real-time
interaction (video banking), ATMs integrated with smart phones, website customization,
37 biometric services, and digital currency. These services can contribute to an increasing
adoption of online services.
39 & 2015 Elsevier Ltd. All rights reserved.

41
63
43
65
45
1. Introduction the choice to acquire and use a new invention or innovation
67
[6]. According to Saleem and Higuchi [7], technology adoption
47
In recent years, technology has increasingly been emp- is the main reason why developing countries are yet to
69
loyed in the delivery of services. Service delivery technology develop; old technology produces less quantity, low quality
49
has become the vital operating elements of today's organiza- product, and short life span. They import old technology due
71
tions that help to reduce costs and improvements in the to less investment in technology, risk aversion, additional cost
51
overall efficiency of operations [1–5]. Technology adoption is of assessment, and lack of appropriate consultant for selection
73
of appropriate technology. Innovation is not based on the R &
53
D alone but in some firms, it is also by the use of innovative
n 75
Corresponding author. product and process from external source [8]. The paradigm
55 E-mail addresses: samsonyusufu@yahoo.com (S. Yusuf Dauda), shift in the way companies interact with their customers, this
jxlee@snu.ac.kr (J. Lee). 77
1 has been led by the proliferation of technology-based systems
57 Department of Science Nasarawa State Polytechnic, P.M.B. 109 Lafia,
Nasarawa State, Nigeria. mostly in the service industry and the banking sector are the
79
59 http://dx.doi.org/10.1016/j.is.2015.04.006
0306-4379/& 2015 Elsevier Ltd. All rights reserved. 81
61

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 2 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]
1 most beneficial. They are developing a number of alternate
delivery channels with a view of attracting tech-savvy custo-
3 mers, improving customers' expectations and ensuring custo-
mer loyalty [9]. The need to explain the user acceptance of
5 new technologies and the factors influencing the acceptance
of such technologies is increasing [10]. According to Martins
7 et al. [11], the banking sector has been using information
services not only to run internal business activities and to
9 promote products, but also to provide main services to
their customers. Thus, the better use of the numerous new
11 information services available in the market is a challenge
facing this sector. Zhu and Chang [12] argued that Information
13 technology is the basis of technology-based services; consu-
mers may be concerned with the usefulness and ease of use of
15 a technology-based service before they decide to avail them-
selves to try it. The effective service delivery on behavioral
17 factors may result in customer loyalty (which impacts their
future utilization of online banking patterns) in relation to
19 other factors just like it has on customer satisfaction [13].
According to Martins et al. [11] adopters of internet and online
21 banking have a lower propensity to leave the bank and have
increased banking activity, acquire more products, and main-
23 tain higher asset and liability balances [5]. Information and
communication technology, competition, deregulation and
25 globalization have forced banks to balance the goals of out-
reach and sustainability, characterized based on the services
27 the banks offer to customers across the globe and in numer-
ous channels [13,14]. Banks have largely implemented service
29 delivery technology as a way of augmenting the services
traditionally provided by bank personnel, but technology-
31 based services could only improve corporate performance
through consumer acceptance or adoption [15,16,12]. Thus,
33 online banking may be the instigator of this new environment
and the prime mover in terms of providing the potential
35 solution for bank's survival in the near future [17–20,4,21].
The level of online banking adoption will directly impact on
37 the degree to which the customers are satisfied, in terms of
the behavioral factors [19]. Several studies analyze online
39 banking service adoption or acceptance when the services are
already deployed and they look at factors that influence
41 customers' adoption and use or intention to use those
services. However, despite the increasing discussion about
43 online banking service adoption in the literature, no attempt
has been made to look at consumers' preference with regards
45 to future online banking service adoption. Based on the
aforementioned, the following questions have been raised
47 by the current study: what is the preference structure and
willingness of the customers to pay for the future online
49 banking services? Which future online banking service pro-
motion policy can be applied base on the heterogeneous
51 population in Nigeria that can best promote an increasing
online banking service adoption?
53 According to Miltgen et al. [22], the isolated impacts of
technical, social, and risk factors on intention to accept IT
55 have limit the ample view of different factors that organiza-
tions trying to succeed with IT implementation have to
57 carefully address in order for the target users to accept the
IT under investigation. Jayawardhena and Foley [23] identi-
59 fied the benefit of increasing the customer base, because
using multiple distribution channels (branch networks, Inter-
61 net banking, mobile banking, etc.) amplifies market coverage
Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
by enabling different products to be targeted at different 63
demographic segments. Thus, with a larger customer base,
banks can profit from marketing and communication, with 65
the possibility of mass customization for each group of
clients, offering innovative products [11]. This is an important 67
issue because many organizations today are saturated with
mass automation and homogenized products and services. In 69
the customer view, there is an increase in the autonomy with
less dependency on the branch banking and consequently 71
less time and effort. According to Miltgen et al. [22], IT is
becoming increasingly complex and crucial for business 73
operations, thus making the issue of acceptance an important
challenge in IT implementation. Despite impressive advances 75
in technology capabilities, the problem of underutilization of
IT, especially for more radical technologies, is still present 77
[24].This study uses a conjoint based discrete choice model-
ing with stated preference data to construct the banking 79
customers' behavior corresponding to the future online bank-
ing service preference of Nigerian banking customers. The 81
objective of this research is to use the conjoint and discrete
choice analysis to investigate and test the variance of 83
consumers' behavior in relation to future online banking
products/service preferences and willingness to pay – that 85
will reveal a basic background for potential pricing themes
and funding directions for banks in the developing countries 87
with specific empirical analysis of the Nigeria banking
industry. This study seek to help banks to understand the 89
type and nature of future online banking services that can be
well accepted by the customers and to create the right 91
policies and actions to attract customers to use these services.
In addition, it is in the banks' and clients' interest to direct 93
their communication from bank branches to online channels
in order to be more productive and cost-effective. 95
This paper is organized as follows: the next section
presents the theoretical background and justification of 97
attributes selection, followed by the research methodol-
ogy. Then, we present the experimental design and data 99
collection, followed by the empirical analysis and findings.
Finally, the interpretation of the findings and both theore- 101
tical and practical implications are described. This paper
concludes by presenting the research limitations and 103
proposing avenues for future research.
105
2. Literature review
107
2.1. Adoption models
109
The growth of information services and IT acceptance and
use has been studied using various models and varied 111
conclusions have been drawn from such studies that offer
new insights at both the individual and organizational levels 113
with focus on a country or a set of countries [15,25,10]. Each
of the several models that have been proposed in the 115
literature has the same dependent variable, use or intention
to use [11], but with various antecedents to understand 117
acceptance of technology. Technology acceptances are infor-
mation service theories that model how users come to accept 119
and use a specific technology. These theories suggest that
when users are presented with a new technology, a number 121
of factors influence their decision about how and when they
will use it. Many authors have studied different aspects of 123
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]
1 new technology acceptance from a variety of theoretical
perspectives explaining the relationship between user beliefs,
3 attitudes, and intentions, including Theory of Reasoned
Action (TRA – [26]), Technology Acceptance Model (TAM –
5 [27]), Theory of Planned Behavior (TPB – [28]), Model of PC
Utilization (MPCU – [29]), Motivational Model (MM – [30]),
7 TPB, a hybrid model combining constructs from TAM and
TPB (C-TAM-TPB – [31]), Innovation Diffusion Theory (IDT –
9 [32–35]), Social Cognitive Theory (SCT – [36]), and Unified
Theory of Acceptance and Use of Technology (UTAUT – [15]).
11 In each of these theories, behavior is viewed as the result of a
set of beliefs about technology and a set of affective responses
13 to the behavior. The most popular theoretical models are TRA,
TPB, TAM, IDT, and UTAUT.
15 The most employed model in the study of information
technology adoption is the TAM [12]. This model examines
17 the causal linkages between two key beliefs and the beha-
vioral attitudes and intentions of users. TAM was designed to
19 predict information technology acceptance and use on the job,
in which perceived usefulness and perceived ease of use are
21 the main determinants of the attitudes [27]. Miltgen et al. [22]
argued that, despite the simplicity and contentiousness of
23 TAM, it has proven particularly useful in studying the intent to
accept new IT in a wide variety of contexts, such as across US
25 companies [15], among college students shopping online [37],
in regards to internet banking [38], e-procurement [39], or
27 electronic toll collection service [40]. TPB focuses more on the
perceived behavioral control, which is viewed in terms of the
29 perceived ease or difficulty of performing the behavior [28].
Both the TAM and TPB models were based on TRA. The TRA
31 model is one of the most important theories of human
behavior drawn from social psychology where attitudes and
33 subjective norms are considered as the determinants of
behavior. The theory proposes that beliefs influence attitudes
35 which in turn lead to intentions and then consequently
generate behaviors [26]. The IDM separates the adopters into
37 categorizes as: innovators, early adopters, early majority, late
majority, and laggards. Innovation must be widely adopted in
39 order to self-sustain. “Individuals are seen as possessing
different degrees of willingness to adopt innovations, and
41 thus it is generally observed that the portion of the population
adopting an innovation is approximately normally distributed
43 over time” [33]. In addition, he proposes that four main
elements influence the spread of a new idea: the innovation
45 itself, communication channels, time, and a social system. In
this regard, an organization decision to adopt an innovation or
47 not relies on organizational and market forces. The UTAUT
model which can explain as much as 70% of the variance in
49 intention [15], postulates that four constructs act as determi-
nants of behavioral intentions and use behavior: performance
51 expectancy, effort expectancy, social influence, and facilitating
conditions. In addition, the model also posits the role of four
53 key moderator variables: gender, age, experience, and volun-
tariness of use. According to Miltgen et al. [22], while the TAM
55 and IDT perspectives focus almost exclusively on beliefs about
the technology and the outcomes of using it, UTAUT includes
57 other beliefs that might influence behavior, independent of
perceived outcomes. Recently, researchers have increasingly
59 turned to testing and applying UTAUT to explain technology
adoption. Examples are: online bulletin boards [41], Web-
61 based learning [42] and instant messengers [43].
Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
3
2.2. Bank service product portfolio 63
Banking service product quality has been found by 65
previous studies to plays an important role in determining
customers' perceptions of overall banking service quality. Jun 67
and Cai [44] explained that banking product quality is
primarily associated with product variety and diverse fea- 69
tures. According to Yang, Jun, and Peterson [45], online
customers are more inclined to patronize firms which offer 71
a substantial variety of services. The primary reason for this
choice is that it is more likely that their diverse needs can be 73
fulfilled. This is especially the case for desired services that
are not widely distributed or unavailable at physical outlets 75
[46]. Thus, a key to gaining customer satisfaction and loyalty
is providing a mix of offerings preferred by target customers. 77
Strieter et al. [47] observed that one of the most important
developments in banking is the increased emphasis on 79
marketing a wide variety of financial services. In terms of
internet shopper satisfaction, Cho and Park [48] have identi- 81
fied “variety of products” as one of the seven key dimensions
that influence internet shopper satisfaction. In the same vein, 83
in terms of the banking, Dixon [49] argues that the key to
getting more customers from the banks through the online 85
service is not the attraction of the internet itself but the
products offered to the customers. Lepkowska-White and 87
Page [50] also pointed out that one of the important
ingredients for developing consumer value in online com- 89
panies is a suitable selection of products/services.
Another rationale for customer use of the internet is 91
convenience. Latimore et al. [51] discovered that when
possible, 87% of internet banking customers prefers to com- 93
plete their transactions at one site. Aliyu [13] considered
convenience to be an influential factor for the use of online 95
banking and that there is a direct relationship between
technology and behavioral (external) factors in the adoption 97
of online banking. For instance, numerous online banking
customers wish to pay their bills electronically and automa- 99
tically, view and print their monthly bank statements, and
purchase stocks, insurance, and other financial offerings. For 101
this reason, companies with wide product lines may be able
to attract a large number of customers to their sites. Also, 103
introducing new forms of products/services to the market-
place appeals to customers whose needs are unfulfilled by 105
existing offerings. Malarvizhi [52] argued that the relationship
of convenience and service delivery via online banking is the 107
ability of online banking to meet users' needs using the
different feature availability of the services. Therefore, a key 109
to gaining customer satisfaction is providing a wide range of
products/services and diverse features in the format required 111
by customers. In line with this, results of Aliyu [13] shows that
convenience and security have strong evidence of customer 113
satisfaction via online banking as the mediator linking the
relationship between online banking and customer service 115
delivery. Therefore, with the advent of internet technology,
present banking customers can have unlimited access to 117
financial information and enjoy a wider range of choices in
selecting financial institutions and competitive products. The 119
key driving force in attracting new customers and enhancing
customers' satisfaction is subtly differentiating quality levels in 121
terms of diverse features of bank products and their timely
introduction to the marketplace [53]. 123
 4 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Table 1 63
Q7 Variable definition for the future bank services.
3 65
Variable name Description Definition Distribution of part-worth

5 Services Variable 67
Rank Rankings of the cards given 1–3, repeated
7 by the respondents 69
ATM services ATMCCT Cardless ATMs (Dummy ‘1’ if it is with capacity; ‘0’ otherwise Normal
variable for ATMs with
9 capacity to perform 71
cardless transaction)
11 Video banking RTINT Real time Interaction ‘1’ if it is Available; ‘0’ otherwise Normal 73
(Dummy variable for Video
technology integrated with
13 most banking channels) 75
Mobile Banking MOBW Mobile wallet (Dummy ‘1’ if it is Available; ‘0’ otherwise Normal
15 variable for mobile wallet 77
services)
Security Services BIOS ‘1’ if it is Available; ‘0’ otherwise Normal
17 ATMISP ATMs smartphone ‘1’ if it is Integrated; ‘0’ otherwise Normal
79
integration (Dummy
19 variable for mobile-based 81
pre-staging of ATM
transactions)
21 83
OCERT Occupational certification ‘1’ if it is Available; ‘0’ otherwise Normal
(Dummy variable for
23 flexible, authentication and 85
transaction security
25 solutions) 87
Internet Banking DCUR Digital currency (Dummy ‘1’ if it is Available; ‘0’ otherwise Normal
variable for Bitcoin
27 services) 89
WEBCUS Website Customization ‘1’ if it is Possible; ‘0’ otherwise Normal
29 (Dummy variable for 91
customers chance to
personalize each page of
31 the website) 93
COSTa Transaction cost (cost per 50; 100; 150 (Nigerian Naira ¼
N/online or wire transfer) Log-normal
33 online or wire transfers) 95

35
2.3. Attributes and attribute levels
37
Nine future online services have been defined as attributes
39 Q3 to establish the choice experiments for the conjoint survey in
the current study. All the attributes are treated as dummies
41 with the values of one if available and zero otherwise. The
characteristics of those attributes and their levels are pre-
43 sented in Table 1. First, the price is a very important factor in
determining the WTP of the consumer. According to Aliyu
45 et al. [54], for consumers to use new technologies, it must be
reasonably priced relative to alternatives. Otherwise, the
47 acceptance of the new technology may not be viable from
the standpoint of the customer. In contrast, results of Aliyu
49 [13] indicated that cost has no direct effect of customer service
delivery via online banking. However, most researchers
51 believe that the price level is always one of the most
important factors that will influence the evaluation result by
53 customers [55,56]. In this research, the levels of price are
constructed hypothetically, but based on real prices of the CBN
55 of Nigeria 2013 revised guide to bank charges.2
57
2
The detail of revised price guide of CBN can be obtained at http://
59 www.cenbank.org/out/2013/fprd/circular%20to%20all%20banks%20and%
20discount%20houses.%20the%20rrvised%20guide%20to%20bank%
61 20charges.pdf.
97
Secondly, cardless ATM access is defined due to the fact
that ATM service is the most popularly used channel of 99
transaction mostly in areas where there is low internet
penetration. Cardless ATM access will allow customers to 101
securely perform transactions at the ATM channel without
the use of a card using one-time SMS PIN or biometric 103
authentication [57–60]. The system includes an ATM and a
net Work each configured to communicate with a mobile 105
user device and the provider system, either a proxy-based
transaction completion (where mobile numbers, one-time 107
PINs or voucher numbers are entered in place of card
swiping), the use of newer quick response codes(QR codes), 109
or near-field communication (NFC) technology, through
which a bank's customers have access to a full range of 111
ATM services including cash withdrawal, money transfers,
debt payments, bill transactions, and counter top-ups. Also, 113
non-customers can have access to bill payment, electronic
check purchase, and redemption and mobile top-up with- 115
out the use of a card.
Third, real-time interaction service was included as one of 117
the future service attributes (e.g. video-enabled mobile
phones, web conferencing, online chat, TV banking, 3D video 119
banking, and video tellers). According to D'Angelo and Little
[61], factors such as navigational and visual characteristics 121
such as video are critical features of a website. Valentine [62]
reported that First Capital Bank of Texas launched ten 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]
1 interactive video teller Kiosks that provides customers with
live, remote teller services at four of the banks seven branches
3 located throughout western Texas. Other examples include
the National Australia Bank, which provides online loan
5 processing through video chat; the SNS Bank Netherlands,
which provides video web conferencing for instructing cus-
7 tomers through multimedia presentations; and Bank Sabadell
Spain, which provides video-enabled phone banking for
9 addressing customer queries.
Fourth, mobile wallet is defined due to the fact that
11 mobile phones, especially smartphones, are becoming one
of the most important channels for banking due to easy
13 accessibility and great importance for mobile and social
customers. The expanded use of smartphones has increased
15 demand for m-banking services [63,64] and is likely to have
significant effects on the market [65]. Mobile wallet is a
17 system that securely stores users' payment information and
passwords for numerous payment methods and websites.
19 The technology can be provided as a turnkey, or a hosted
solution enabling financial institutions, retailers, payment
21 service providers, merchants, banks, telecom operators, utility
providers, or other businesses to offer their customers smart
23 and practical branded services. It can be used in conjunction
with mobile payment systems that allow customers to pay
25 for purchases with their smartphones, and can also be used
to store loyalty card information, tokens, and digital coupons.
27 Unlike debit or credit cards, transactions using an electronic
wallet are carried out off-line without the direct involvement
29 of financial intermediaries and the burden of these institu-
tions' high fixed costs [66,67]. In addition, Van Hove [68]
31 argues that even though electronic wallets, are frequently
compared to debit cards, it should rather be compared to
33 cash since the aim is to provide consumers and merchants
with an electronic payment instrument that could handle
35 small transactions cost effectively. Different mobile wallets
use different technologies and offer unique rewards to
37 customers. Some mobile wallets use NFC to transfer payment
information to a vendor's POS terminal, e.g., Google Wallet,
39 PayPal, or Square Wallet.
The fifth, sixth and seventh attributes will be clarified
41 by the security and privacy of online transaction. Many
researchers have emphasized the importance of security in
43 online banking [44,69–73,45], no customer wants to reveal
their personal information, such as credit card numbers,
45 due to the growing rate of online fraud. In addition, Aliyu
et al. [74] identified security as an important characteristic
47 from a customer's perspective on the adoption of innova-
tion. The current study defines three security measures as
49 future service attributes: biometrics sensors, ATMs inte-
grated with smartphones, and occupation certification.
51 Biometrics sensor-related services include: ATM/e-bank-
ing face recognition systems, voice recognition services,
53 optical sensor, finger print recognition, and name recognition
services. According to McGarr [75], the fact that biometric
55 systems require two forms of input for identity verification,
including biometric input along with a personal identification
57 number (PIN), makes the systems more powerful. In fact, the
augmented interest in biometric technology is led by antici-
59 pated decrease of technology costs, improved technical
quality of the systems and socio-political pressures for better
61 security-related controls [76]. However, user resistance to
Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
5
utilize such pervasive technology is still an issue [76]. 63
According to Bankston [77], while the technology goes back
years and has been used in highly sensitive institutions such 65
as defense and nuclear facilities, the proliferation of electronic
data exchange has generated new demand for biometric 67
applications that can secure electronically stored data and
online transactions. Various researchers have identified the 69
crucial need to inform biometric technologies implementa-
tion with various factors affecting its acceptance [78,76]. 71
However, only a few authors have discussed biometric
systems from a consumer acceptance perspective [78,22]. 73
The integration of ATMs with smartphones involves a
mobile-based pre-staging of transactions (where the mobile 75
device replaces the ATM screen and keyboard); transactions
are started on a smartphone, where a one-time code is 77
received and typed into the ATM, releasing the cash. Accord-
ing to Dabholkar [79], there is greater control when the 79
customer is in direct contact with technology such as internet
banking and m-banking. Joseph and Stone [80] added that 81
empowering customers by providing them with the option of
using technology-based service delivery systems may be a 83
relatively inexpensive way to maintain customer loyalty. For
example, at the 2013 International Consumer Electronics 85
Show, Diebold, Inc., a software and technology firm in North
Canton, Ohio, unveiled an ATM that is designed to work with 87
smartphones.
Occupational certification is a comprehensive and flex- 89
ible authentication and transaction security solution that
allows financial organizations to meet the needs of an 91
escalating threat landscape and tailor mitigation methods
to different customers' risk profiles based on their online 93
banking activity [81] today and as threats and opportu-
nities evolve in the future. Even though the perception and 95
behavioral response of end users are important considera-
tions when designing systems that employ digital identi- 97
ties [82] as issues of privacy, security and online identity
management are frequently a source of concern to con- 99
sumers [83]. Solutions may range from easy-to-deploy,
software-based out-of-band (OOB) strong authentication 101
solutions for low-risk online banking activities to digital
transaction signing solutions for validating online transac- 103
tions and use of digital certificates.3 In addition to this, an
SMS notification for designated online transactions is 105
provided along with the digital certification.
Finally, the eighth and ninth attributes will be clarified 107
by the internet banking service of which digital currency
and website customization are incorporated as future 109
banking service attributes. The attraction of banks to
online banking is fairly obvious: diminished transaction 111
costs, easier integration of services, improvements in the
overall efficiency of operations, interactive marketing 113
capabilities, and other benefits that boost customer lists
and profit margins [79,44,1,80]. Additionally, web banking 115
services allow institutions to bundle more services into
single packages, thereby luring customers and minimizing 117
overhead. Banks look to the web as a way of maintaining
their customers and building loyalty [80]. 119
121
3
See more on digital certificates at http://www.netpnb.com/web/
L001/webpages/digicertuserguide.pdf. 123
 6 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Digital currency (bitcoin) was introduced as open-source characteristics are assumed to be homogeneous, assuming 63
software in 2009 as an innovative payment network and a that each εnj is independently and identically distributed
3 new kind of money. Bitcoin uses peer-to-peer payment (IID) type I extreme value distribution [90]. Also assuming 65
system technology to operate with no central authority or independence of irrelevant alternatives (IIA) [89] i.e. the
5 banks; managing transactions and the issuing of bitcoins is relative odds of choosing ‘i’ over ‘k’ are the same no matter 67
carried out collectively by the network and everyone can what other alternatives are available or what the attributes
7 take part. Steadman [84] explains it as crypto-currency, of the other alternatives are. 69
because it uses cryptography to control the creation and V ni
e
9 transfer of money. Conventionally, the capitalized word P ni ¼ P V nj
ð3Þ 71
Bitcoin refers to the technology and network, whereas je

11 lowercase bitcoin refers to the currency itself. Rank-ordered logit model is an extension to multi- 73
Website customization service allows banks to give nomial logit model in the sense that an individual n states
13 customers a chance to personalize each page of the   75
their ranking of alternativer n ¼ r n1 ; r n2 ; ……; r nJ , as the
website, with the customers deciding which features will descending order of preference [87]. Then, facing the
15 show up on the screen and where. Many researchers maximization problem, the order of choice is made if: 77
believe that banks need to customize their websites to

17 meet customer preferences [61,85,86,44]. According to U ðr n1 Þ 4 U ðr n2 Þ 4 … 4U r nJ ð4Þ 79
Dabholkar [79], there should be flexibility in the design
19 of the technology to allow customers to make changes 
 J1 eβxðrk Þ 81
Pr U ðr 1 Þ 4 U ðr 2 Þ 4… 4 U r J ¼ ∏ PJ ð5Þ
during the transaction. Joseph and Stone [80] added that k¼1 m¼k
eβxðrm Þ
21 this also raises design issues with the technology itself, 83
including ease of access, ease of use, and provision of where k denotes the order of alternative ranked at kth by
23 sufficient menu options on ATM and internet banking the respondent. In the current study the utility is specified 85
services. Personalized experience can also help customers as in Eq. (6) and the variable definitions and descriptions
25 understand their needs and get recommendations based are given in Table 1. 87
on their age, income group etc., or create tag-based U ijt ¼ βATMCCT ATMCCTþ βRTINT RTINTþ βMOBW MOBW þ βBIOS BIOS
27 interactions that help users retrieve information about 89
their past online banking activities. Although, customiza- þ βATMISP ATMISP þ βOCERT OCERT þ βDCUR DCUR
29 tion cannot be priced explicitly, it does help in bolstering þ βWEBCUS WEBCUS þ βCOST COSTþ εijt ð6Þ 91
customer retention and improving loyalty, thus adding to
31 the bank's profitability from individual accounts. The RUM with part-worths interacting with demo- 93
graphics and individual characteristics is also specified in
33 the current study as follows: 95
3. Methodology U ijt ¼ βATMCCT ATMCCTþ ðβRTINT þ βIFMRT IFM þ βIURT IUÞRTINT
35 97
þ ðβMOBW þ βIAMW IAþ βIFMMW IFMÞMOBW
3.1. Random utility model


37 þ βBIOS þ βIABS IA BIOS þ βATMISP þ βIFMASP IFM 99
The random utility model is used as a theoretical basis 
þ βICASP IC þβBAASP BA ATMISPþ ðβOCERT þβIFMOC IFM
39 for analyzing consumer preferences using discrete choice 101
models [87]. Regarding the banking services, the customers þ βIUOC IUÞOCERT þ ðβDCUR þ βIADC IAþ βPIBDC PIB


41 have their own valuation on the future banking service þ βBCDC BCÞDCUR þ βWEBCUS þβIUWC IU þ βIBWC IB 103
attribute and their own characteristics. Thus, we assume 
þ βIOWC IO WEBCUSþ ðβCOST þ βAGECO AGEþ βWKECO WKE
43 that each customer perceives the utility associated with 105
each attribute of the future banking services and chooses þ βBTCO BTþ βPCHCO PCH þβPMBCO PMBþ βBBCO BB
45 the service with the greatest possible perceived utility þ βBACO BAÞCOST þεijt ð7Þ 107
[88,89]. This utility is decomposed into two parts in random
47 utility model: deterministic (observable by the researcher) IFM is the internet familiarity of the respondents, IU is 109
and stochastic (unobservable by the researcher). the internet usage frequency of the customer, IA is the
49 internet access of respondents, IC is the internet access of 111
U nj ¼ V nj þ εnj ð1Þ
respondents at a cafe, BA is the respondents whose major
51
 bank is Bank A, PIB is percentage of IB of the respondents, 113
U nj ¼ V nj þ εnj ¼ V X nj ; Sn þ εnj ð2Þ
BC is the respondents whose major bank is Bank C, IB is the
53 where U nj ¼utility that decision maker n obtains from respondents who use IB, IO is the internet access of 115
alternative j; V nj ¼representative utility/deterministic part respondents at the office, AGE is the age of the respondents,
55 of consumers utility, Sn ¼attributes of decision maker; WKE is the working experience of the respondents, BT is 117
εnj ¼error term. the bank transaction frequency of the respondents, PCH is
57 the respondents having a PC at home, PMB is the percen- 119
3.2. Multinomial logit model tage of m-banking of the respondents, BB is the respon-
59 dents whose major bank is Bank B. According to Savage and 121
In the setting of multinomial logit model the degrees of Waldman [91], these interaction terms relax the restriction
61 influence of attributes of product/service and individual of fixed coefficients that are specified in Eq. (6). 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]] 7

1 3.3. Random coefficient model (mixed logit) where the parameters estimates β^ k and β^ c are the true 63
parameters and using the corresponding elements in the
3 The random coefficient model gives space for random asymptotic variance-covariance matrix [94]. 65
variation in taste, relaxation from the IIA property, and
5 correlation among unobserved factors in repeated choices 67
4. Experimental design and data collection
avoid biased utility results [92]. The stochastic

term
 can be
7 decomposed into two additive parts εn ¼ ηn þσ n , a sto- 69
4.1. Conjoint experiment
chastic part ðηn Þ that is correlated over alternatives, having
9 its own distribution that is defined by the researcher 71
The conjoint questionnaire was developed in English.
depending on underlying parameters and observed data
There are five choice sets; each choice set has three
11 relating to the alternative j and the individual n and 73
alternatives. Each respondent answered the choice ques-
another stochastic part ðσn Þ that is independently and
tionnaire by ranking alternatives in each choice set. The
13 identically distributed over alternatives and individuals. 75
choice sets are orthogonally designed using the SPSS,
Accordingly, the utility of a customer n from alternative j
therefore avoiding individual effect. A pilot survey of over
15 can be defined as follows: 77
55 samples in Abuja the Federal Capital and Lagos State
U nj ¼ X nj βn þεnj ð8Þ the former Capital was conducted. This was followed by
17 79
the main survey conducted in five States in Nigeria by
where unknown parameter βn , which comprises a vector
face-to-face interviews led by research assistants. The
19 of coefficients of bank service attributes X nj , allows a 81
areas surveyed were evenly spread throughout in all
variation in tastes with respect to the customers. In the
regions (North, Central, and South regions). These even
21 current study the random coefficients is defined similar to 83
disparities made sure the responses received represent all
Eq. (6). The difference is that all the coefficients βn are
the preferences of citizens in the country. The other Q4
23 assumed to have their own distribution in the population. 85
aspects of the population such as gender and age are also
eV nj ðβÞ distributed evenly to make the survey stratified in terms of
25 Lnj ðβÞ ¼ PK ð9Þ 87
eV nk ðβÞ the age and sex ratio according to the population of the
k¼1 4
country. A sample of the conjoint questionnaire is given
27 V nj ðβÞ is the observed part of the utility. 89
in Appendix A (Table 2).

29 91
4.2. Data collection
3.4. Willingness to pay
31 93
The data for this study was obtained through a conjoint
Given the consumer income and budget constraints
33 survey using the current banking customers. Of 1700 ques- 95
willingness-to-pay (WTP) is what the consumers are pre-
tionnaires that were distributed, a total of 1412 responses
pared to spend on a certain goods or services. Following
35 were received corresponding to an initial response rate of 97
Chaudhuri et al. [93], WTP for banking services in this study
(83.06%). After discarding invalid responses, we had 1291
is related to a set of underlying service characteristics that
37 responses left of which 51.74% respondents are male and 99
combine to produce a single, separable index of service
48.26% female. The other questionnaires are invalid because
utility to consumers. WTP for future bank services is affected
39 they were either left “blank”, “not fully filled in” or “only one 101
by existing bank service accessibility, use experience, percep-
number was written” in the choice sets. The age range is
tion, affordability, and ability to pay, together with custo-
41 from 18 to 82 of which (39.43%) are in the age range of 18– 103
mers' level of awareness of bank services as well as internet
29, (35.09%) are between 30 and 39, (19.29%) are between 40
use experience. According to Bliemer [94], when a utility
43 and 49, (4.26%) are between 50 and 59, and (1.94%) are 60 or 105
function is linear in parameters and attributes, the WTP for a
older. For analytical purposes the age range of respondents is
one unit improvement in that attribute is the ratio of its
45 divided into – below 30 (39.43%) and at least 30 (60.57%). 107
marginal utility to the marginal utility of price:
The demographics of the respondents can be shown in Fig. 1.
47 wk ¼ βk =βc ð10Þ For the entire sample, there is low percentage of those 109
educated from primary to secondary school (13.71%). Simi-
where βk is the parameter for attribute k and βc is the cost
49 larly, there was also low percentage of those with master and 111
parameter. In the more general case of a nonlinear utility
doctoral degrees (14.87% master and 3.02% doctorate). The
function, the WTP of attribute k is defined as:
51 majority of respondents are those with high school diplomas 113
∂g j =∂X jk or technicians with technical or professional training in
wk ¼ ð11Þ
53 ∂g j =∂X jc colleges, polytechnic, or technical training centers (35.48%), 115
followed by those with a university education (32.92%). Two
Two main methods for determining confidence inter-
55 educational groups was focused: Education 1 refers to those 117
vals for WTP measures in the case of fixed coefficients are
educated up to diploma level (49.91%), while the rest
the Krinsky and Robb method (K&R) and Delta methods
57 Education 2 belongs to the ones having highest intellectual 119
[94]. For the delta method, in this case used in this study,
has been defined by Bliemer [94] as:
59 121
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi



ffi
4
Details of the Nigerian population and sex ratio can be obtained at
seðW^ k Þ ¼ 1 var β  2wk cov β ; β þ w2 var β ð12Þ https://www.cia.gov/library/publications/the-world-factbook/geos/ni.
k k c c
61 βc k
htmlThe WorldFactbookwww.cia.gov. 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 8 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Table 2 63
Demographic statistics.
3 65
Characteristics of study subjects N¼1291

5 Characteristics Percentage Characteristics Percentage Characteristics Percentage Characteristics Percentage 67

7 Age range (18–64 yrs) Monthly income(USD$) Transaction frequency Bank cheque usage 69
Age1 ( o30 yrs) 39.43 Low (0–315) 49.57 1–2/week 76.76 0/week 59.72
Age2 ( Z30 yrs) 60.57 High ( Z 316) 50.43 Z3 23.24 1/week 29.59
9 Gender Marital status No. of credit cards Z 2/week 10.69 71
Female 48.26 Single 47.71 0 48.95 Internet access
11 Male 51.74 Married 47.10 1 34.55 No 19.67 73
Education Others 5.19 Z2 16.60 Yes 80.33
Edu1 (r Diploma) 49.19 Family size No. of debit cards Internet familiarity
13 Edu2 (Z Bachelor) 50.81 Small (1–5) 52.05 0 10.07 Not at all 19.44 75
Occupation Big ( Z6) 47.95 1 62.97 Fairly 7.36
15 Unemployed 16.81 Major bank Z2 26.96 Averagely 24.48 77
Civil Servant 51.43 Bank A 24.86 Credit card usage Well 24.86
Businessman 14.33 Bank B 22.54 0/week 49.81 Very well 23.86
17 Academic 9.84 Bank C 23.16 1/week 27.03 Internet usage
79
Others 7.59 Other banks 29.43 Z 2/week 23.16 Never use 18.59
19 Working experience Deposit size Debit card usage 1/week 9.45 81
Exp1 (0–5 yrs) 52.83 Low (0–315) 81.87 0/week 10.38 2–3/week 14.25
Exp2 (6–10 yrs) 23.00 High ( Z 316) 18.13 1/week 48.72 4–5/week 17.04
21 83
Exp3(Z 11 yrs) 24.17 Z 2/week 40.90 Every day 40.67

23 85
90 5. Results and discussion
80
25 70
87
60 5.1. Rank-ordered logit model
50
27 40
89
30 The total observations for ranked-ordered logit estimation
29 20 are 6455. This study uses the LIMDEP program to estimate the 91
10
0 rank-ordered logit model. The results of rank-ordered logit
Age Gen Edu Inc Dep Major Bank
31 estimation are shown in Table 3 that includes the parameter 93
group A group B group C group D
estimates, asymptotic t-statistics, willingness-to-pay of con-
33 Fig. 1. Demographics of respondents. sumers for each attribute of the future banking services and 95
relative importance of the attributes. This study interprets the
35 level of education from bachelor to doctorate (50.81%). The parameters as marginal utility that is as a partial derivative, 97
monthly income level of respondents was also categorized which means the change in utility for a unit increase in the
37 into 2 groups: the low income group earning up to $315 US variable [91]. In addition, the significant or confidence interval 99
dollars per month (49.57%) and high income group, above of the WTP has been calculated using the delta mesa method
39 $315 US dollars per month (50.43%). In the same way, the from Eq. (12) and their significant level with the asymptotic 101
monthly deposit of the respondents was also categorized t-statistics is also presented in Table 3.
41 into 2 groups and majority of the respondents were found to From Table 3, consumers have high utility with increase in 103
have very low deposits (81.87%) ranging from $0 to 315. As access to ATMs with capacity for cardless transaction, real-
43 for the major banks, majority of the respondents use other time-interaction services, mobile wallets, ATMs integrated 105
banks (29.43%), while the smallest percentage used Bank B with smartphones, occupational certification, digital currency
45 (22.54%), with Bank A (24.86%) and Bank C (23.16%) in the and website customization services. On the other hand, 107
middle. Customers' use debit cards more (40.90%) as com- customers have less utility with increase in service fees. This
47 pared to credit cards (23.16%) and bank cheque (10.69%). finding is in line with other researchers [55,56,54], and in 109
51.43% of the respondents were civil servants of which contrast with results of Aliyu [13] that indicated that cost has
49 52.83% of them have very low experience (0–5years) and no direct effect of customer service delivery via online bank- 111
almost 52.05% have a small family size (1–5 people). 80.33% ing. The biometric service was not significant in this model
51 of the respondents have internet access. About 19.44% of the showing that; even though the biometric services are intro- 113
respondents are not familiar with the internet at all, while duced their utility does not change. This finding is similar to
53 23.86% are very well familiar with the internet. 18.59% have [76,78] who pointed out the issue of user resistance to utilize 115
never used internet, while almost 40.67% of the respondents such pervasive technology and crucial need to inform bio-
55 use the internet every day. This is reasonable because the metric technologies implementation with various factors 117
internet penetration rate for Nigeria as at 2012 is 32.8% affecting its acceptance. Also from Table 3, the result of the
57 according to the World Bank 2012 estimate.5 relative importance reveals that as a whole, the consumers 119
rate the service fee and real-time-interaction services as the
59 most important future banking services (20.33%) and 121
(18.65%) respectively. The least important is biometric service
61 5
http://databank.worldbank.org/data/views/reports/tableview.aspx. (5.58%). If we translate it into willingness to pay (WTP), based 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]] 9

1 Table 3 63
Q8 Rank-ordered logit estimation results.
3 65
Variable matching Variable Coeff. β Std. Err. t-value WTPa WTP t-value Rel. imp.

5 Cardless ATMs ATMCCT 0.3259nnn 0.0876 3.722 61.14nnn  2.577 12.43 67

Real time Interaction RTINT 0.4890nnn 0.0982 4.982 91.73nnn  2.902 18.65
7 Mobile wallet MOBW 0.2046n 0.0944 2.168 38.38n  1.853 7.80 69
Biometrics sensors BIOS 0.1464 0.1012 1.448 N/A 0.000 5.58
ATMs smartphone integration ATMISP 0.1759nnn 0.0419 4.201 32.99nnn  2.721 6.71
9 Occupational certification OCERT 0.2015nnn 0.0529 3.809 37.79nnn  2.605 7.68 71
Digital currency DCUR 0.2209nnn 0.0433 5.097 41.43nnn  2.925 8.42
11 Website Customization WEBCUS 0.3248nnn 0.0695 4.673 60.92nnn  2.837 12.39 73
Service cost COSTa  0.0053nnn 0.0015  3.571 N/A N/A 20.33

13 N ¼6455. Asterisk nnn indicates significant level at 99.9% (p-value o 0.001). LRI: ρ ¼0.431. Unit of WTP is ¼
N 1 the exchange rate was $1 ¼ ¼
N 160. 75
a
Willingness to pay (WTP).
15 77

17 on this model, customers are willing to pay a high premium
price of about ¼N¼N 90.73 per month for real-time-interaction
19 services. Similarly, they are willing to pay high price for
cardless ATMs and website customization ¼ N 61.14 per month
21 and ¼N 60.92 per month respectively. For the other services,
their willingness to pay is reasonably high ¼
N 41.43 per month
23 for digital currency service, ¼N 38.38 per month for mobile
wallet service, ¼
N 37.79 per month for occupational certifica-
25 tion service, and ¼
N 32.99 per month for ATMs integrated with
smartphones. Based on WTPs, policy makers and bank
27 management can establish their general priorities on which
future online banking service to promote that will bring
29 about increase in adoption of online services and subse-
quently lead to customer's satisfaction.
31
5.2. Model with interaction SDCs
33
The result of the interaction of demographics and other
35 individual's characteristics (income, deposit, education,
etc.), which allows a systematic test variation and enables
37 the heterogeneity to be incorporated into the part-worth is
as presented in Table 4. Accordingly, the model made its
39 interaction with eight future online banking service attri-
butes. The sign of the mean values of the coefficients are
41 the same as those in Table 3.The order of the importance of
attributes is also similar except for their magnitudes,
43 which were slightly higher for the interaction model.
From Table 4, customers still rate the service fee as the
45 most important attribute (20.23%) followed by the real-time
interaction services (19.02%). The least-preferred future
47 online service is still the use of biometric sensors (5.08%).
However, the results are not homogenous for all respon-
49 dents; with the interaction terms we can see an improve-
ment in the significance level of the biometric sensor service
51 since it was not significant in the first model. First, an
individual's relative utility increases with increasing access
53 to biometrics services (βBIOS ¼0.4368) but it is likely that the
more a customer has access to the internet the lower their
55 preference for the use of biometric sensors βIABS (0.3873).
Meaning that, the consumer preference for the use of
57 biometric sensors is changing from those that have internet
access to those that do not have access to the internet.
59 Instead of providing general information about the popula-
tion's lack of preference to the biometric sensor services as
61 in rank-ordered logit, these flexible results enable us to
categorize the preference levels depending on demographic 79
and individual characteristic groups. As a result, consumers
are willing to pay ¼ N 25.12/month for biometrics services if 81
made available to them. Secondly, an individual's relative
utility decreases when the price increases (βCOST ¼ 0:0143), 83
but it decreases more among the individuals who have more
working experience (βWKECO ¼  0:0001) or have a higher 85
transaction frequency per week (βBTCO ¼ 0:0005). It may
even decrease much more among those individuals whose 87
major banks are Bank A (βBACO ¼ 0:0024) and Bank B
(βBBCO ¼  0:0026), while it decreases less in the case of 89
those who are of the higher age group (βAGECO ¼ 0:0003),
have a PC at home (βPCHCO ¼ 0:0040), or have a higher 91
percentage of m-banking use (βPMBCO ¼ 0:286E  4). Clearly,
such categories of individuals do not care much about the 93
cost. Meaning that, younger people with higher working
experience are more sensitive to price and also those who do 95
not have PCs and access the internet at public centers. Thus,
consumers' choice is based on their utility level. It is likely 97
that at certain pricing levels, those that are of a higher age
group, have a PC at home, and are already using m-banking 99
will chose to use the future online services while those that
have more frequent bank transactions, have more work 101
experience, and use Bank A or Bank B will choose not to.
Thirdly, an individual's relative utility increases with an 103
increase in access to real-time interaction services
(βRTINT ¼ 0:7832), but it is more steady for those who have 105
lower frequency of internet usage (βIURT ¼  0:0226) and
even much more for those that are less or not very familiar 107
with the internet (βIFMRT ¼  0:0715). Meaning that, real-
time interaction services are more effective when they target 109
those who have a lower frequency of internet usage and
those in the rural and remote areas that are less familiar with 111
the internet. As a result, consumers are willing to pay as
much as ¼ N 94.03/month for real-time interaction if made 113
available to them. Fourthly, an individual's relative utility
increases with an increase in access to mobile wallet services 115
(βMOBW ¼ 0:5757), but decreases among individuals that are
more familiar with the internet (βIFMMW ¼  0:0681) and 117
even much more among individuals that already have access
to the internet (βIAMW ¼ 0:2218). However, consumers are 119
willing to pay ¼N 37.35/month for the mobile wallet service if
made available to them. Fifth, an individual's relative utility 121
increases with an increase in access to ATM smartphone
integration services (βATMISP ¼ 0:3635). However, the relative 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 10 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Table 4 63
Model with interaction.
3 65
Variable Coefficient Std. Err. t-Value MWTPa Rel. Imp.

5 Cardless ATMs ATMCCT βATMCCT 0.3017nnn 0.0884 3.414 60.27 12.19 67

Real time Interaction RTINT βRTINT 0.7832nnn 0.1081 7.244 94.03 19.02
7 Mobile wallet MOBW βMOBW 0.5757nnn 0.1071 5.373 37.35 7.56 69
Biometrics sensors BIOS βBIOS 0.4368nnn 0.1125 3.882 25.12 5.08
ATMs S. phone integration ATMISP βATMISP 0.3635nnn 0.0602 6.037 33.80 6.84
9 Occupational. Cert. OCERT βOCERT 0.3334nnn 0.0702 4.748 38.01 7.69 71
Digital currency DCUR βDCUR 0.4382nnn 0.0646 6.779 43.70 8.84
11 Web Customize WEBCUS βWEBCUS 0.5314nnn 0.0821 6.469 62.05 12.55 73
Service cost COSTa βCOST  0.0143nnn 0.0019  7.502 N/A 20.23
Age and Cost AGECO βAGECO 0.0003nnn 0.400E  4 6.253
13 Work Exp. & Cost WKECO βWKECO  0.0001n 0.0001  2.002 75
Bank Trans. & Cost BTCO βBTCO  0.0005nn 0.0002  2.606
15 PC at Home & Cost PCHCO βPCHCO 0.0040nnn 0.0006 6.493 77
Inter. Acc. & Mob. Wallet IAMW βIAMW  0.2218nn 0.0896  2.475
Inter. Acc. & Bio. Sensor IABS βIABS  0.3873nnn 0.0578  6.696
17 Inter. Acc. & Dig. Currency IADC βIADC  0.1770nn 0.0591  2.997
79
Inter. Fam. & Real Time Inter. IFMRT βIFMRT  0.0715nnn 0.0172  4.164
19 Inter. Fam. & Mob. Wallet IFMMW βIFMMW  0.0681nnn 0.0210  3.240 81
Inter. Fam. & ATM S. phone Integration IFMASP βIFMASP  0.0505nnn 0.0134  3.769
Inter. Fam. & Occ. Cert. IFMOC βIFMOC  0.0761nnn 0.0170  4.475
21 83
Internet Usage and Real Time Interaction IURT βIURT  0.0226n 0.0106  2.124
Inter. Use & Occ. Cert. IUOC βIUOC 0.0227n 0.0113 2.002
23 Inter. Use & Web Customize IUWC βIUWC  0.0368nnn 0.0101  3.661 85
Inter. Banking & Web Customize IBWC βIBWC  0.0579n 0.0271  2.135
25 Per. of Inter. Banking & Digital Currency PIBDC βPIBDC  0.0022nnn 0.0007  3.425 87
Per. of Mob. Banking & Cost PMBCO βPMBCO 0.286E  4nnn 0.749E  5 3.817
Inter. at Office & Web Customize IOWC βIOWC  0.1828nnn 0.0485  3.773
27 Inter. at Café & ATM S. phone Integration ICASP βICASP  0.2201nn 0.0729  3.020 89
Bank A & ATM S. phone Integration BAASP βZBASP  0.1179n 0.0582  2.025
29 Bank C & Digital Currency BCDC βGTBDC  0.1067n 0.0483  2.207 91
Bank B & Cost BBCO βFBCO  0.0026nnn 0.0006  4.398
Bank A & Cost BACO βZBCO  0.0024nnn 0.0007  3.504
31 93
NB: N¼ 6455, Asterisk *** indicates significant level at 99.9% (p-value o 0.001). LRI: ρ ¼0.439. At the time of survey, the exchange rate was $1 ¼ ¼
N 160.
a
33 Currency unit of WTP is ¼N 1 (Nigerian naira). 95

35 utility decreases among customers that are more familiar
with the internet (βIFMASP ¼  0:0505) and even much more
37 among those who access the internet at public centers such
as internet cafes (βICASP ¼  0:2201) and among those indi-
39 vidual whose major bank is Bank A (βBAASP ¼  0:1179). This
implies that, ATM smartphone integration services are more
41 effective when they target those in the rural and remote
areas that are less familiar with the internet; those that
43 access the internet at centers other than internet cafes, such
as at home or at the office or using mobile phones or
45 smartphones; and those customers whose major bank is
not Bank A. As a result, consumers are willing to pay ¼ N 33.80/
47 month for ATM smartphone integration services if made
available to them. Sixth, an individual's relative utility
49 increases with an increase in access to occupational certifica-
tion service (βOCERT ¼ 0:3334) but the more familiar an
51 individual is with the internet, the lower their preference
for this service βIFMOC (0.0761). However, as a customer use
53 the internet frequently it is more likely that they will have a
higher utility with access to occupational certification ser-
55 vices βIUOC (0.0227). This reveals that familiarity with a future
online service alone does not guarantee a customer's pre-
57 ference for adoption of that service when deployed and if he
does not use internet frequently, which shows that experi-
59 ence of internet use has a great impact on a customer's
preference for future online services. Specifically, in the case
61 of occupational certification, it implies that the more
customers use the internet the more they understand the 97
risks of not having good security due to the increase in
internet fraud, which makes them want to have improved 99
and better security. This finding is in line with the findings of
Sathye's [85] study on the adoption of IB by Australian 101
consumers, which found that difficult in use and security
were the two most important reasons that customers did not 103
want to use the service. Liu and Arnett [86] also added that
system use is one of the four factors when considering the 105
major ingredients for success of a website. As a result,
customers are willing to pay ¼ N 38.01/month for the occupa- 107
tional certification service if made available to them. Seventh,
an individual's relative utility increases with an increase in 109
access to digital currency services (βDCUR ¼ 0:4382). How-
ever, the relative utility of a customer decreases among those 111
who have access to the internet (βIADC ¼  0:1770), who
have a high percentage of IB (βPIBDC ¼  0:0022), and whose 113
major bank is Bank C (βBCDC ¼  0:1067). Meaning that,
digital currency services are more effective when they 115
targets those who have no access to the internet, who have
a low percentage of IB; and whose major bank is not Bank C. 117
As a result, consumers are willing to pay ¼ N 43.70/month for
digital currency services if made available to them. Finally, 119
though in general people have a high preference for website
customization (βWEBCUS ¼ 0:5314), individuals that have a 121
high frequency of internet usage, those that are already
doing IB, and those that access the internet at the office have 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]] 11

1 Table 5 63
Estimation results of the random coefficient model.
3 65
Variables Mean (b) of B Variance (w) of B

5 Mean Std. Err. t-value Std. Div. Variance Std. Err. t-value 67
nnn nnn
7 ATMCCT 1.0138 0.0994 10.2 0.4782 0.2287 0.0651 7.35 69
RTINT 1.1243nnn 0.1141 9.85  0.0039 1.5E  05 0.0723  0.05
MOBW 1.2161nnn 0.1052 11.56 0.2808 0.0788nn 0.1149 2.44
9 BIOS 1.0957nnn 0.1156 9.48 0.0103 0.0001 0.0826 0.13 71
ATMISP 0.4539nnn 0.0498 9.12 0.0106 0.0001 0.0820 0.13
11 OCERT 0.5242nnn 0.0613 8.56  0.3680 0.1354nnn 0.0800  4.6 73
DCUR 0.0413 0.0562 0.74 0.5827 0.3395nnn 0.0652 8.94
WEBCUS 0.7108nnn 0.0936 7.59 0.0453 0.0021 0.0673 0.67
13 COSTa  0.6715nnn 0.1563  4.3 1.0822 1.1712nnn 0.1108 9.77 75

15 N ¼19,365; Asterisk (n) marks the significance at 5% (|t| 41.65); LR chi2 (10) ¼ 806.80, Log likelihood ¼  6592.9407, LRI: ρ ¼0.672. 77

17 a lower marginal utility regarding website customization
services (βIUWC ¼ 0.0368, βIBWC ¼  0.0579, and βIOWC ¼
19 0.1828). This implies that website customization services
are more effective when they target those that have a lower
21 frequency of internet usage, those that are not doing IB, and
those that access the internet at their homes or internet cafes
23 or using mobile phones or smartphones. As a result, con-
sumers are willing to pay a very good amount for this
25 service: up to ¼ N 62.05/month for website customization
services if made available to them.
27 The factor of high frequency in the interaction terms is
internet familiarity, showing its importance as a strong
29 influence on future online banking service adoption. The
more familiar a person is with the internet the more they
31 want to have access to it, and subsequently will use the
internet more frequently. In this way, they can be moti-
33 vated to adopt the future online services based on their
experience and habit of frequently using the internet.
35
5.3. Random coefficients model
37
The random coefficient model that relaxes the restric-
39 tions of the ML model is analyzed by assigning the
distribution for the part-worth in the RUM [95]. For the
41 estimation, maximum-likelihood estimation with a sample
size of 19,365 was used. This research follows the estima-
43 tion method used by Train [89], Kim [96], and Lee et al.
[97]: StataSE 12 (64-bit) was used to do the estimate and
45 the estimated means and their variance are shown in
Table 5, in which almost all the results are significant at
47 5% except the mean estimate of the parameter DCUR.
The estimation assumes a log-normal distribution for
49 the transaction cost. The means of coefficient estimates in
Table 5 have the same signs and same order of significance
51 as those of Table 3, except for MOBW, BIOS and DCUR
where the significant levels changed. In general, the
53 variances in random coefficient estimates are relatively
smaller compared to their means except in the cost and
55 digital currency attributes. Hence it shows that the varia-
tion in the consumer price and service attributes is clearly
57 captured, showing either extreme heterogeneity or het-
erogeneous preferences of the consumers depending on
59 type of future banking service. In addition, not all esti-
mates of variance are statistically significant, showing that
61 customers are homogeneous with regards to some future
online services. The WTP for each attribute was also 79
calculated along with the significant or 95% confidence
interval, the result along with the asymptotic t-statistics is 81
also presented in Table 6.
First from Table 5, the mean coefficient of digital 83
currency service is small (0.0413) and not significant in
this model showing that there are some few minorities 85
who actually do not care, so even though the digital
currency service is introduced, their utility level does not 87
change. This finding is in contrast to the finding of the
rank-ordered logit model, which shows that the whole 89
population has a positive preference for digital currency.
The large variance (0.3395) reveals the heterogeneity in 91
preference. Second, the mean coefficient of website cus-
tomization services is large (0.7108) showing that a 93
majority of customers like the website customization
services which supports the finding of the rank-ordered 95
logit model. The insignificantly small variance (0.0021)
reveals that they are highly homogeneous in their pre- 97
ference to this service. Third, the mean coefficient of ATMs
integrated with smartphones services is large (0.4539) 99
showing that quite a reasonable amount of customers like
the ATMs integrated with smartphones services which 101
supports the finding of rank-ordered logit result. Similarly,
the insignificant variance (0.0001) shows that they are 103
extremely homogeneous regarding this service. Fourth, the
mean coefficient for occupational certification service is 105
also high (0.5242). This supports the findings of rank-
ordered logit result showing that a majority of customers 107
like the occupational certification service, but the high
variance (0.1354) reveals the heterogeneous in their pre- 109
ference. Fifth, the mean coefficient for ATMs with capacity
for cardless transactions is very large (1.0138). This also 111
supports the findings of rank-ordered logit result showing
that a majority of customers like the ATMs with capacity 113
for cardless transactions, while the high variance (0.2287)
shows that they are highly heterogeneous in their pre- 115
ference. Sixth, the mean coefficients for real-time interac-
tion services are also very large (1.0138), which supports 117
the findings of rank-ordered logit result showing that a
majority of customers like the real-time interaction ser- 119
vices. However, the small insignificant variance (1.5E  05)
reveals that they are highly homogeneous with regards to 121
this service. Seventh, the mean coefficient for biome
tric sensor services is very large (1.0957) showing that a 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 12 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Table 6 63
Willingness to pay for random coefficient estimates.
3 ¼)
65
Variables WTP (US$) WTP (N

5 Mean [95% Conf. Interval] Mean [95% Conf. Interval] 67

7 Lower Upper Lower Upper 69

Cardless ATMs ATMCCT 1.5099 1.0425 1.9772 241.584 166.8 316.352
9 Real time interaction RTINT 1.6743 1.1675 2.1811 267.888 186.8 348.976 71
Mobile wallet MOBW 1.8111 1.2526 2.3697 289.776 200.416 379.152
11 Biometrics sensors BIOS 1.6318 1.1635 2.1000 261.088 186.16 336 73
ATMs smartphone integration ATMISP 0.6760 0.4512 0.9007 108.16 72.192 144.112
Occupational certification OCERT 0.7807 0.5173 1.0441 124.912 82.768 167.056
13 Digital currency DCUR 0.0615  0.0861 0.2092 9.84  13.776 33.472 75
Website customization WEBCUS 1.0585 0.6940 1.4231 169.36 111.04 227.696
15 Service cost COSTa N/A N/A N/A N/A N/A N/A 77
At the time of estimation $1 ¼ ¼
N 160.
17 79

19 majority of customers like the biometric sensor services.
This finding reveals the shortcoming of the rank-ordered
21 logit model, which shows that the utility of the consumers
does not change even with the introduction of biometric
23 sensor services as it was not significant in the rank-
ordered logit model. However the small insignificant
25 variance (0.0001) shows that preferences are extremely
homogeneous. Finally, the mean coefficient for mobile
27 wallet services is also very large (1.2161) showing that a
majority of customers like mobile wallet services which
29 supports the finding of the rank-ordered logit model. The
small variance (0.0788) shows that they are not too
31 different in their opinion about the mobile wallet services.
From Table 6, the result of the WTPs for the random
33 coefficient model shows that majority of consumers are
willing to pay only ¼
N 9.84/month for digital currency service,
35 which is quite true since it was not significant in this model
compared to ¼ N 108.16/month for ATMs with smartphone
37 integration, ¼N 124.91/month for occupational certification,
and ¼N 169.36/month for website customization. On the other
39 hand, customers are willing to pay very high amount for
other services (approximately ¼ N 289.78 for mobile wallet
41 services, ¼
N 267.89/month for real-time interaction services,
and ¼N 241.58/month for cardless ATMs). In addition, custo-
43 mers are willing to pay ¼N 261.09/month for biometric service,
which is in contrast to the rank-ordered logit model, as it
45 was not significant in that model. These amounts are quite
high compared to the current payment levels of ¼ N 150/
47 transaction for online transfers, ¼
N 200/transaction for inter-
bank transfers using an ATM and online bill payment, ¼ N 5 for
49 commission on turn over (COT), and ¼ N 4 for alert charges.
51 6. Managerial implications
53 Based on the above results, several implications have
been drawn for management. First, the critical need for
55 managers to focus on promoting these future online ser-
vices as they have strong impact on customer's satisfaction
57 or dissatisfaction depending on service features or demo-
graphic and characteristics groups has been emphasized.
59 However, relative to other services, the service fee and real-
time interaction are the most important services. First, in
61 line with other researchers, Koenig-Lewis et al. [98], Luo
[99], and Laukkanen et al. [100] young people who use 81
online banking regularly are early adopters of innovative
technologies that will eventually filter through to older age 83
groups and with increased word of mouth, it can lead to a
network effect that will reach critical mass adoption. 85
Managers can take advantage of this and introduce services
that add entertainment for youths or extra convenience in 87
terms of ease of usage [101,102], e.g., digital wallets, real-
time interaction (video banking), ATMs integrated with 89
smartphones, website customization, and digital currency.
Second, since the ATM is the most common channel used 91
by most customers for transactions, ATM integrated ser-
vices have a better chance of increasing adoption, especially 93
in developing countries where there is low access to the
internet, low internet banking and m-banking adoption. 95
Third, even though website customization services cannot
be priced explicitly, they can help in customer retention and 97
improve satisfaction. Since majority are interested in this
service, Banks can concentrate on developing a customized 99
website with tag-based interactions that help users retrieve
information about their past online banking activities for 101
the majority while providing a standard website for other
groups. Fourth, promoting digital currency may not be an 103
effective policy. Meanwhile, Bank A, Bank C, and other
banks can be the chief promoters of digital currency 105
targeted at those with less access to the internet and lower
percentage of IB use. Business models based on P2P NFC 107
should be promoted to help create new payment-driven
revenue streams. Fifth, there should be a policy for promot- 109
ing comprehensive, flexible, and additional authentication
transaction security solutions that will allow the banks to 111
meet the needs of an escalating threat landscape and tailor
mitigation methods to different risk profiles. Sixth, even 113
though the biometric service is preferred less, Bank man-
agers can introduce this by providing various options in 115
terms of differentiated services since majority that like the
service are not very heterogeneous in their preference. 117
Clearly, Bank B, Bank C, and other banks can be the chief
promoters of these services, which should be targeted at 119
customers who are less familiar with the internet and
access the internet at home or the office or using mobile 121
phones or smartphones. In addition, other bank strategies
to encourage potential adopters like offering incentives for 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]] 13

1 adoption, providing customers valued-added promotion
programs, and aggressive marketing could lead to increas-
3 ing adoption.
5 7. Conclusions
7 In order to attain global networking and increased
efficiency of the banking industry, especially in developing
9 countries, one of the most important tasks of management
is promoting the adoption and use of e-banking systems
11 for fast and efficient delivery of services that will lead to
increased sales and market share at the same time meet
13 customer satisfaction, attract new customers, and retain
existing ones. The aim of this paper therefore is to
15 construct the banking customer's behavior corresponding
to future online service preference. This study has used
17 conjoint analysis and stated preference methods using
discrete choice modeling to study the technology adoption
19 pattern regarding consumers' preference for potential
future online banking services in the Nigerian banking
21 industry. As a foundation for policy making, this research
has incorporated heterogeneity into the models using a
23 random coefficient model as well as interaction between
the primary bank services attributes and individual demo-
25 graphics and characteristics and the results are consistent
with the conventional rank-ordered logit model. They are
27 also consistent with previous studies and have proved to
be suitable under the conditions of low market penetra-
29 tion and in areas where there is no market information, as
in the case of new products that have not yet entered the
31 market which reveals future technologies that can foster
an increased adoption of online banking services. Policies
33 to promote smart and practical branded services especially
self-services and promoting a universal adoption of
35 e-banking systems will contribute to increasing adoption
of online services, as customers can enjoy making in-store
37 payments quickly and securely without having to locate
their credit or debit card. Findings from the study are
39 expected to have a profound impact on and important
implications for product and service selection, marketing
41 campaigns, and operating strategies for Nigerian banks
and other financial organizations. Empirically, the affecting
43 vectors reflecting both exogenous and endogenous factors
that have an impact on a bank customer's choice of service
the current online services that may impact on their 63
perception to adopt the future online services. Similarly,
studies specifically discussing factors that might prevent 65
the adoption of future online banking services were also
considered beyond the scope of this study. Finally, our data 67
is limited to potential banking customers only. However,
there are several areas that have the potential to deliver 69
additional and relevant Insights; an example is microfi-
nance institutions that serve low-income markets. For 71
further studies, studies on consumers' preference for future
online services that undertake a comparative analysis of 73
developed and developing economies using different mod-
els and approaches could produce meaningful insights into 75
the behavior and attitudes of customers. According to
Crabbe et al. [103] as cited in Shaikh and Karjaluoto [64], 77
“cross-cultural and transnational studies would enable
researchers to determine how specific social and cultural 79
characteristics of a society influence the adoption of tech-
nologies and services among its members”. In addition, a 81
research that will investigate the behavior of smartphone or
tablet users in relation to the adoption of future online 83
services will be interesting. Following Shaikh and Karja-
luoto [64], since this subset of consumers that have smart- 85
phone or tablet have convenient access to the internet; the
basic computer skills required for conducting various online 87
activities such as online shopping and online account man-
agement; and the basic computer skills and the necessary 89
technology infrastructure to conduct m-banking, relies on
the integration of wireless and internet technologies [104], 91
they may be ready for the adoption of the future online
services. 93
Appendix Sample conjoint experiment card 95
97
Part 1. Assuming that you can accrue possible benefits from future 99
online Banking services including ATMs with capacity to perform
cardless transaction, real time interaction services, mobile wallets, 101
biometrics sensors, ATMs integration with smartphones, occupation
certification, digital currency, and website customization/
personalization by paying some cost as transaction cost. Please rank 103
the types of future online Banking Services from 1 to 3 (1¼ most
preferred service, 3¼ least preferred service) from the list of 5 105
hypothetical options provided below.

45 can be referred to in future banking studies. More sig- Please rank from most preferred to least preferred future Banking
107
nificantly, the level of importance of each factor has also services from the four options given here.
47 been revealed, and the preference structure is likely to 109
provide a well-established background for reference in
49 future studies on banking service marketing as well as in Questionnaire 1: Future Online Banking Services 111
policy making areas.
51 Some limitations from the current study offer opportu- Attributes of banking services Service A Service B Service C 113
nity for additional research. First, the current study focused
53 on consumers' preference on future online services, but 1. ATMs with capacity to With No No 115
perform cardless transaction capacity capacity capacity
online services are vast in scope, comprising detailed
2.Real time interaction Available Available Not
55 analysis from the supply side such as cost investment, services available 117
infrastructure, technology, innovation, and efficiency. Incor- 3. Mobile wallet Not Available Available
57 porating all these aspects of electronic banking into future available 119
studies will be useful for a more in-depth analysis. Second, 4. Biometrics sensors Not Not Available
available available
59 the core of the current study is acceptance or adoption of 5.ATMs integration with Not Not Integrated
121
future online banking services, so it excluded some beha- smartphones integrated integrated
61 vior intensions of customers based on experience and use of 6. Occupation certification Available 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 14 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]

1 Not Not [19] A. Musiime. Bank clients' perception of information technology 63

available available usage, service delivery and customer satisfaction: reflections on
7. Digital currency Available Not Available uganda's banking sector, in: Proceedings of the International Trade
3 available & Academic Research Conference (ITARC), London, 2010.
65
8. Website customization/ Not Possible Possible [20] H.A. Kadir, N. Rahmani, R. Masinaei, Impacts of service quality on
5 personalization possible customer satisfaction: study of online banking and ATM services in 67
9. Cost (transaction cost) ¼
N 50 ¼
N 50 ¼
N 150 Malaysia, Int. J. Trade Econ. Finance 2 (1) (2011) 1–9.
Rank from 1 to 3 (1¼ most () () () [21] T. Markku, Impact of future trends on banking services, J. Internet
7 preferred service & 3¼ least Bank. Commer. 17 (2) (2012) 271–295. 69
preferred service) [22] C.L. Miltgen, A. Popovič, T. Oliveira, Determinants of end-user accep-
9 tance of biometrics: integrating the Big 3 of technology acceptance 71
Note: Assume that all the other attributes, besides the nine with privacy context, Decis. Support Syst. 56 (2013) 103–114.
proposed here, remain the same. [23] C. Jayawardhena, P. Foley, Changes in the banking sector – the case
11 ofInternet banking in the UK, Internet Res. 10 (1) (2000) 19–30. 73
[24] E.W. Baker, S. Al-Gahtani, G.S. Hubona, Cultural impacts on accep-
13 tance and adoption of information technology in a developing 75
country, J. Glob. Inf. Manag. 18 (3) (2010) 35–58.
[25] I. Im, S. Hong, M.S. Kang, An international comparison of technology
15 Appendix A. Supporting information adoption: testing the UTAUT model, Inf. Manag. 48 (1) (2011) 1–8. 77
[26] M. Fishbein, I. Ajzen, Belief, Attitude, Intention and Behavior: An
Introduction to Theory and Research, , 1975.
17 Supplementary data associated with this article can be 79
[27] F.D. Davis, Perceived usefulness, perceived ease of use, and user
found in the online version at http://dx.doi.org/10.1016/j.is. acceptance of information technology, MIS Q. 13 (3) (1989)
19 2015.04.006. 319–339. 81
[28] I. Ajzen, The theory of planned behaviour, Org. Behav. Hum. Decis.
Process. 50 (2) (1991) 179–211.
21 [29] R.L. Thompson, C.A. Higgins, J.M. Howell, Personal computing: 83
toward a conceptual model of utilization, MIS Q. 15 (1) (1991)
References
23 124–143. 85
[30] F.D. Davis, R.P. Bagozzi, P.R. Warshaw, Extrinsic and intrinsic
[1] J. Alstad, Use your service edge to your online advantage, Am. Bank motivation to use computers in the workspace, J. Appl. Soc.
25 167 (2002) 1–3. Psychol. 22 (14) (1992) 1111–1132. 87
[2] R. Byers, P. Lederer, Retail bank service strategy: a model of [31] S. Taylor, P.A. Todd, Assessing IT usage: the role of prior experience,
27 traditional, electronic, and mixed distribution choices., J. Manag. MIS Q. 19 (2) (1995) 561–570. 89
Inf. Syst. 18 (2) (2001) 133–156. [32] E.M. Rogers, Bibliography on the Diffusion of Innovations, , 1962.
[3] T. Oliveira, M.F. Martins, Literature review of information technol- [33] E.M. Rogers, Diffusion of Innovations, 5th ed. Free Press, New York,
29 ogy adoption models at firm level, Electron. J. Inf. Syst. Eval. 14 (1) London, Toronto, Sydney, 1995. 91
(2011) 110–121. [34] E.M. Rogers, Diffusion of Innovations, New York, 5th ed. The Free
[4] D. Thulani, K. Njanike, C. Manomano, L. Chiriseri, Adoption and use Press, 2003.
31 of sms/mobile banking services in Zimbabwe: an exploratory study, [35] G.C. Moore, I. Benbasat, Integrating diffusion of innovations and
93
J. Internet Bank. Commer. 16 (2) (2011) 149–167. theory of reasoned action models to predict utilization of information
33 [5] M. Xue, L.M. Hitt, P. Chen, Determinants and outcomes of Internet technology by end-users, in: J. Kautz, K. Pries-Hege (Eds.), Diffusion 95
bankingadoption, Manag. Sci. 57 (2) (2011) 291–307. and Adoption of Information Technology, 1996, pp. 132–146.
Q5 [6] l.B.H. Hall, B. Khan, Adoption of technology, New Economy Hand- [36] D.R. Compeau, C.A. Higgins, Computer self-efficacy: development
35 book, University of California, 2002. of a measure and initial test, MIS Q. 19 (2) (1995) 189–211. 97
[7] A. Saleem, K. Higuchi, Policy perspective analysis of technology [37] S. Ha, L. Stoel, Consumer e-shopping acceptance: antecedents in a
adoption in industries, Int. J. Smart Home 8 (2014) 1. technology acceptance model, J. Bus. Res. 62 (5) (2009) 565–571.
37 99
[8] A. Arundel, C. Bordoy, & M. Kanerva. Neglected innovators: how do [38] S.Y. Yousafzai, G.R. Foxall, J.G. Pallister, Explaining internet banking
innovative firms that do not perform R&D innovate. Results of an behavior: theory of reasoned action, theory of planned behavior, or
39 analysis of the Innobarometer, 9, 2007. technology acceptance model? J. Appl. Soc. Psychol. 40 (5) (2010) 101
[9] A. George, G.G. Kumar, Impact of service quality dimensions in 1172–1202.
internet banking on customer satisfaction, DECISION 41 (1) (2014) [39] A.M. Gamal, Predicting e-procurement adoption in a developing
41 73–85. 103
country: an empirical integration of technology acceptance model
[10] K.K. Kapoor, Y.K. Dwivedi, M.D. Williams, Rogers' innovation adop-
and theory of planned behaviour, Ind. Manag. Data Syst. 110 (3)
43 tion attributes: a systematic review and synthesis of existing 105
(2010) 392–414.
research, Inf. Syst. Manag. 31 (1) (2014) 74–91.
[40] C.D. Chen, Y.W. Fan, C.K. Farn, Predicting electronic toll collection
[11] C. Martins, T. Oliveira, A. Popovič, Understanding the Internet
service adoption: an integration of the technology acceptance
45 banking adoption: a unified theory of acceptance and use of 107
model and the theory of planned behavior, Transp. Res. C: Emerg.
technology and perceived risk application, Int. J. Inf. Manag. 34
Technol. 15 (5) (2007) 300–311.
(1) (2014) 1–13.
47 [12] D.H. Zhu, Y.P. Chang, Investigating consumer attitude and intention
[41] J.T. Marchewka, C. Liu, K. Kostiwa, An application of the UTAUT 109
toward free trials of technology-based services, Comput. Hum. model for understanding student perceptions using course man-
Behav. 30 (2014) 328–334. agement software, Commun. IIMA 7 (2) (2007) 93–104.
49 [42] C.-M. Chiu, E.T.G. Wang, Understanding Web-based learning con-
111
[13] A.A. Aliyu, T. Rosmain, J. Takala, Online banking and customer
service delivery in Malaysia: data screening and preliminary tinuance intention: the role of subjective task value, Inf. Manag. 45
51 findings, Procedia-Soc. Behav. Sci. 129 (2014) 562–570. (3) (2008) 194–201. 113
[14] M. Al-Hawari, N. Hartley, T. Ward, Measuring banks automated [43] C.P. Lin, B. Anol, Learning online social support: an investigation of
service quality: a confirmatory factor analysis approach, Mark. Bull. net-work information technology based on UTAUT, Cyber Psychol.
53 Behav. 11 (3) (2008) 268–272. 115
16 (2005) 2005.
[15] V. Venkatesh, G.B. Davis, F.D. Davis, M.G. Morris, User acceptance of [44] M. Jun, S. Cai, The key determinants of internet banking service
55 information technology: toward a unified view, MIS Q. 27 (3) quality: a content analysis, Int. J. Bank Mark. 19 (7) (2001) 276–291. 117
(2003) 425–478. [45] Z. Yang, M. Jun, R.T. Peterson, Measuring customer perceived online
[16] J. Jasperson, P.E. Carter, R.W. Zmud, A comprehensive conceptuali- service quality: scale development and managerial implications,
57 zation of post-adoptive behaviors associated with information Int. J. Oper. Prod. Manag. 24 (11) (2004) 1149–1174. 119
technology enabled work systems, MIS Q. 29 (3) (2005) 525–557. [46] S.M. Barcia, Internet pharmacies: all hype with no help, Health
[17] S.M. Nsouli, A. Schaechter, Challenges of the e-banking revolution: Manag. Technol. 21 (4) (2000) 24–25.
59 121
finance and development, Int. Monet. Fund 39 (3) (2002) 231–254. [47] J. Strieter, A.K. Gupta, S.P. Raj, D. Wilemon, Product management
[18] F.L. Anyasi, P.A. Otubu, Mobile phone technology in banking and the marketing of financial services, Int. J. Bank Mark. 17 (7)
61 system: its economic effect, Res. J. Inf. Technol. 1 (1) (2009) 1–5. (1999) 342–354. 123

Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
 S. Yusuf Dauda, J. Lee / Information Systems ] (]]]]) ]]]–]]]
1 [48] N. Cho, S. Park, Development of electronic commerce user-
consumer satisfaction index ecusi) for Internet shopping, Ind.
Manag. Data Syst. 101 (8) (2001) 400–405.
3 [49] M. Dixon, Com madness: 9 must-know tips for putting your bank
online, Am. Community Bank. 8 (6) (1999) 12–15.
5 [50] E. Lepkowska-White, C. Page, Web equity: a framework for building
consumer value in online companies, J. Consum. Mark. 19 (3)
(2002) 231–248.
7 [51] D. Latimore, I. Watson, & C. Maver. The customer speaks: 3300
Internet users tell us what they want from retail financial services.
〈http://www.mainspring.com/research/document/view/1〉, 2099
9 [81] M. Koller, & Y. Rutrell. Security Gets Some Legs: Digital ID Systems
(1215), 00, 2000.
[52] V. Malarvizhi, An analysis on the usage of e-banking services in
11 coimbatore city, IJBEMR 2 (1) (2011) 226–242.
[53] N.P. Mols, The Internet and services marketing-the case of Danish
retail banking, Internet Res. Electron. Netw. Appl. Policy 10 (1)
13 (2000) 7–18.
[54] A.A. Aliyu, R. Tasmin, The impact of information and communica-
15 tion technology on banks' performance and customer service
delivery in the banking industry, Int. J. Latest Trends Finance Econ.
2 (1) (2012) 80–90.
17 [55] A.D. Athanassopoulos, Customer satisfaction cues to support mar-
ket segmentation and explain switching behavior, J. Bus. Res. 47
(2000) 191–207.
19
[56] A. Arbore, B. Busacca, Customer satisfaction and dissatisfaction in
retail banking: exploring the asymmetric impact of attribute
21 performances, J. Retail. Consum. Serv. 16 (2009) 271–280.
[57] M.C. Stinson, J.W. TemplerJr, & D. Clower. U.S. Patent No. 6286756.
Washington, DC: U.S. Patent and Trademark Office, 2001.
23
[58] W.F. MannIII, & J.L. Hirka. U.S. Patent No. 7103576. Washington, DC:
U.S. Patent and Trademark Office, 2006.
25 [59] F.M.I. William, & J.L. Hirka. U.S. Patent No. 7099850. Washington,
DC: U.S. Patent and Trademark Office, 2006.
[60] R. Varadarajan. U.S. Patent Application 13/048,096, 2011.
27 [61] J. D’Angelo, S. Little, Successful Web pages: what are they and do
they exist? Inf. Technol. Libr. 17 (2) (1998) 71–81.
29 [62] L. Valentine, Bankers' hours takes on new meaning: Texas bank is a
firm believer in video banking. but branches will stay, ABA Bank. J.
106 (5) (2014) 33.
31
[63] A.A. Shaikh, Mobile banking adoption issues in Pakistan and
challenges ahead, J. Inst. Bank. Pak. 80 (3) (2013) 12–15.
[64] A.A. Shaikh, H. Karjaluoto, Mobile banking adoption: a literature
33
review, Telemat. Inf. 32 (1) (2015) 129–142.
[65] R. Safeena, H. Date, A. Kammani, N. Hundewale, Technology
35 adoption and Indian consumers: study on mobile banking, Int. J.
Comput. Theory Eng. 4 (6) (2012) 1020–1024.
[66] Z. M’Chirgui, & O. Chanel. The adoption and use of smart card
37
technology in banking: an empirical evidence from the Moneo
electronic purse in France. Advances in Banking Technology and
39 Management: Impacts of ICT and CRM, Hershey, PA: Information
Science Reference, 2008, pp. 70–87.
[67] D.L. Amoroso, R. Magnier-Watanabe, Building a research model for
41 mobile wallet consumer adoption: the case of Mobile Suica in
Japan, J. Theor. Appl. Electron. Commer. Res. 7 (1) (2012) 94–110.
[68] L. Van Hove, Electronic purses in Euroland: why do penetration
43
and usage rates differ? SUERF Studies, (2004/4) (2004).
[69] V.A. Zeithaml, A. Parasuraman, A. Malhotra, A conceptual frame-
45 work for understanding e-service quality: implications for future
research and managerial practice MSI Working Paper series, Report
No. 00-115, , 2001.
47 [70] V.A. Zeithaml, A. Parasuraman, A. Malhotra, Service quality delivery
through Web sites: a critical review of extant knowledge, J. Acad.
49 Market. Sci. 30 (4) (2002) 362–375.
[71] C.N. Madu, A.A. Madu, Dimensions of e-quality, Int. J. Qual. Reliab.
Manag. 19 (3) (2002) 246–258.
51 [72] J. Santos, E-service quality: a model of virtual service quality
dimensions, Manag. Serv. Qual. 13 (3) (2003) 233–246.
53 [73] M.F. Wolfinbarger, M.C. Gilly, eTailQ: dimensionalizing, measuring
and predicting etail quality, J. Retail. 79 (3) (2003) 183–198.
[74] A.A. Aliyu, An exploratory study on adoption of electronic banking:
55 underlying consumer behaviour and critical success factors – case
of Nigeria, Bus. Manag. Rev. 2 (1) (2012) 1–6.
[75] M.S., McGarr. Tuning in biometrics to reduce E-commerce risk
57 Electronic Commerce World, 2000.
[76] A.P. Pons, P. Polak, Understanding user perspectives on biometric
59 technology, Commun. ACM 51 (9) (2008) 115–118.
Please cite this article as: S. Yusuf Dauda, J. Lee, Technology adoption: A conjoint analysis of consumers' preference on
future online banking services, Information Systems (2015), http://dx.doi.org/10.1016/j.is.2015.04.006i
15
[77] K. Bankston. Biometrics: toys or tools? Credit Union Management, 61
2001. 63
[78] C. Morosan, Theoretical and empirical considerations of guests'
perceptions of biometric systems in hotels: extending the technology Q6
acceptance model, J. Hosp. Tour. Res. (2010). (1096348010380601). 65
[79] P. Dabholkar, Technology based service delivery, Adv. Serv. Mark.
Manag. 3 (1994) 241–271.
[80] M. Joseph, G. Stone, An empirical evaluation of US bank customers
67
perceptions of the impact of technology on service delivery in the
banking industry, Int. J. Retail Distrib. Manag. 31 (4) (2003) 69
190–202.
No Longer Hold Users Captive to a Single Browser, PC or Location 71
Internet Week, 2000.
[82] L.A. Jones, A.I. Antón, & J.B. Earp. Towards understanding user
73
perceptions of authentication technologies, in: Proceedings of the
2007 ACM Workshop on Privacy in Electronic Society, ACM,
October, 2007, pp. 91–98. 75
[83] F. Deane, K. Barrelle, R. Henderson, D. Mahar, Perceived accept-
ability of biometric security systems, Comput. Secur. 14 (3) (1995)
225–231. 77
[84] I. Steadman. Wary of Bitcoin? A guide to some other cryptocur-
rencies. Ars Technica. (Retrieved 2014.01.10), 2013. 79
[85] M. Sathye, Adoption of Internet banking by Australian consumers:
an empirical investigation, Int. J. Bank Mark. 17 (7) (1999) 324–334.
[86] C. Liu, K.P. Arnett, Exploring the factors associated with Web site 81
success in the context of electronic commerce, Inf. Manag. 38 (1)
(2000) 23–34.
[87] D. McFadden, Conditional logit analysis of qualitative choice
83
behavior, in: P. Zarembka (Ed.), Frontiers in Econometrics, Aca-
demic Press, New York, 1973, pp. 105–142. 85
[88] H. Varian. Intermediate Microeconomics: A Modern Approach.
Norton & Company. ISBN-10: 0393927024, 1999.
87
[89] K. Train, Discrete Choice Method with Simulation, Cambridge
University Press, Cambridge, New York, 2003.
[90] D. Koh, Ph.D. dissertation, Advances in Demand Analysis and
89
Forecasting of New Products Using Discrete Choice Models and
Empirical Applications, Seoul National University, South Korea,
2007. 91
[91] S.J. Savage, D. Waldman, Ability, location and household demand
for Internet bandwidth, Int. J. Ind. Org. 27 (2009) 166–174.
[92] D. McFadden, K. Train, Mixed MNL models for discrete response, J. 93
Appl. Econom. 15 (2000) 447–470.
[93] A. Chaudhuri, K. Horrigan, J. Flamm, An analysis of the determi- 95
nants of Internet access, Telecommun. Policy 29 (2005) 731–755.
[94] M.C. Bliemer, J.M. Rose, Confidence intervals of willingness-to-pay
97
for random coefficient logit models, Transp. Res. B: Methodol. 58
(2013) 199–214.
[95] D.F. Layton, Random coefficient models for stated preference
surveys, J. Environ. Econ. Manag. 40 (2000) 21–36.
99
[96] Y. Kim, J.D. Lee, D. Koh, Effects of consumer preferences on the
convergence of mobile telecommunications devices, Appl. Econ. 37
101
(7) (2005) 817–826.
[97] J. Lee, Y. Kim, J. Lee, Y. Park, Estimating the extent of potential
competition in the Korean mobile telecommunications market: 103
switching costs and number portability, Int. J. Ind. Org. 24 (2006)
107–124.
105
[98] N. Koenig-Lewis, A. Palmer, A. Moll, Predicting young consumers'
take up of mobile banking services, Int. J. Bank Mark. 28 (5) (2010)
410–432. 107
[99] Y. Luo, Using Internet data collection in marketing research, Int.
Bus. Res. 2 (1) (2009) 196–202.
[100] T. Laukkanen, S. Sinkkonen, M. Kivijärvi, P. Laukkanen, Innovation 109
resistance among mature consumers, J. Consum. Mark. 24 (7)
(2007) 419–427. 111
[101] S.W. Ezzi, A theoretical Model for Internet banking: beyond
perceived usefulness and ease of use, Arch. Bus. Res. 2 (2) (2014)
31–46. 113
[102] C.S. Yu, Consumer switching behavior from online banking to
mobile banking, Int. J. Cyber Soc. Educ. 7 (1) (2014) 1–28.
[103] M. Crabbe, C. Standing, S. Standing, H. Karjaluoto, An adoption
115
model for mobile banking in Ghana, Int. J. Mob. Commun. 7 (5)
(2009) 515–543. 117
[104] X. Luo, H. Li, J. Zhang, J.P. Shim, Examining multi-dimensional trust
and multi-faceted risk in initial acceptance of emerging technolo-
gies: an empirical study of mobile banking services, Decis. Support 119
Syst. 49 (2) (2010) 222–234.