Service sectors have witnessed a rapid shift particularly in the last decade
under the pressure of technology, which is creating new products, services,
market opportunities and developing more information and system-oriented business
and management processes (Liao and Cheung, 2002). In
the world of banking; the development in information technology have had an
enormous effect in development of more flexible payment methods and more efficient
banking services. This technology offers institutions some efficient delivery
channels through which customer banking requirement can be delivered more conveniently
and economically (Akinçi et al., 2004;
Brodie et al., 2007; David et al., 2008). One of the delivery
channels introduced for financial services is Internet Banking (IB) or online-banking.
To use Internet based financial services, consumers not only need to understand
the technology they also need to understand financial services. It is well accepted
that IB is a useful tool in banking system that offers less waiting time and
is more convenient than traditional branch banking (Pikkarainen
et al., 2004). In addition, this new banking system has significantly
lower cost structure than traditional delivery channels. This delivery channel
deserves special attention owing to its enormous potential from the viewpoint
of banks, businesses and retail customers. Bank management have interest in
studies which have investigated the adoption of IB as the results can shed light
on how to better market their Internet banking services and thus, accelerate
the rate of adoption. If the service can more quickly reach a critical mass
of customers (Rogers, 1995), the respective banks
investment in IB could be recouped more quickly.
Financial sector is the spinal cord of sovereign economy of any country. Iran is no exception. The technology adaptation in banking operations in Iran was a few decades behind that of in developed countries. In terms of international and global comparison, Internet usage in Iran is still relatively low and therefore online-banking is still in the infancy stage. Presently in Iran, an important function of e-commerce is the handling of payment over the Internet. Most e-commerce involves the exchange of some form of money for goods and services. Implementation of payment system for e-commerce is still evolving in Iran thus the number of proposal and implementation of payment system currently compete for dominant.
The literature review indicates the lack of non model-based studies to date
has sought to quantify of prominent demographic, behavioral and attitudinal
characteristics as predictor of customer choice in the context of IB in Iran.
To address this lacuna the present study was undertaken. The result could help
bank managers to make informed decisions, thereby providing better services
to their customers and formulate more efficacious strategies to ensure rapid
migration of customers to IB.
Customer adoption: Consumers learn about goods and services by the development
of experiences and the experiences are a major determinant of consumer choice
and preferences (Foxall, 2003). It is revealed that
not only banks customers use but also customers acceptance supports and
determines the success of Internet banking and this has a great influence on
its adoption (Pikkarainen et al., 2004). As time
passes, financial institutions world-wide become more interested in diversifying
their traditional service delivery channels, basically the branch network, which
is known to be associated with high staff and overhead cost. The advent of new
channels has contributed not only to the adoption of multi-channel strategies
by the existing institutions but also the emergence of new forms of financial
businesses such as virtual banks. This trend is evident in the related literature
that several studies have comparatively investigated the current usage levels
and advantages of financial distribution channels (Akinçi
et al., 2004). The advantages of Internet channels and especially
the Internet over traditional branch banking were also underlined. Jayawardhena
and Foley (2000) listed the advantages for banks: cost savings, increased
customer base, mass customization, marketing and communication, innovation and
development of non-core business. In the context of technologically based distribution
channels, such as IB; customers attitudes and motives have generally been
significant for banks managers. If Iranian customers are not adopting IB services,
it may be because they are not aware about such services being available and
the benefit it offers. This concept leads us to different variables to be examined.
The critical question is: What are the main effective factors that influence
the adoption of retail IB services by Iranian customers?
To find an answer to this question, the past researches specifically paramount
study of Lee et al. (2005) enabled us to extract
the variables as well as personal and behavioral characteristics which have
influence on adoption of IB services. Each of the identified measure will be
stated in a set of variables to determine adopters belief in adoption
of the characteristic in question.
Mellat Bank from governmental sector is studied in this research because of
its positive response for conducting the research in bank. Mellat Bank has been
established on April 1980 with a paid of capital of Rls 33.5bn. Currently, the
bank is one of the largest commercial bank in the Islamic Republic of Iran,
ranking among the top 1000 bank of the world and one of the main governmental
bank that have focused on electronic banking services, especially Internet banking
(Economic focus, 2006, 2007).
The history of Internet banking: Innovative banking, since 1981 that
was the introduction to the Automotive Teller Machines (ATMs), has grown aided
mainly by technological developments in the telecommunications and information
technology industry. Early 1990s was the time that Automated Voice Response
(AVR) was introduced to the financial industry which gave the institutions the
ability to launch telebanking services and facilities to their customers. As
technology is revamping the ways that financial services are produced and delivered
(Claessens et al., 2002); with these technological
development, by Intranet proprietary software, banks were able to offer services
to customers who owned personal computers (Sohail and Shanmugham,
2003). NetBank in USA was the first online bank which was formed in 1996
under the name Atlanta Internet Bank and in 1997, second electronic bank; WingSpan
was organized. Other online financial services such as Juniper.com, e-Trade.com
entered the electronic banking environment in 2001 and well Established banks
such as Citibank and Wells Fargo moved into this area of banking by offering
such services to their existing customers in 2001 (Gefen
and Straub, 2000). At recent years e-banking has experienced explosive growth
and has transformed traditional practices in banking dramatically (Gonzalez
et al., 2008).
Over the last 20 years, more of daily life activities are moved online in Iran
and for this, electronic banking services can be important examples of this
trend. The evolution of the electronic banking industry in Iran can be traced
at late 70s. This was the time when two banks installed the first Automatic
Teller Machines (ATM) in Tehran. Later, by the fast developments of technology
in early 90s the Iranian banks upgraded their automation standards and gradually,
with the nationwide growth in Internet connectivity; fundamental platforms of
online data transmission formed. At late 90s a few of Iranian banks began to
implement such services (Economic focus, 2007). Nowadays;
technology runs through every part of the banking business in Iran. IB adoption
in Iran is relatively low and very little research has been carried out to understand
the key adoption determinants. In Iran IB is still new and in infancy stage
and customers are less familiar and often more skeptical toward online banking
transaction due to lack of knowledge and security.
Adoption is the acceptance and continued use of a product, service or idea.
According to Rogers (1995), before customers can adopt
an innovation they must learn about it, which this learning is called the adoption
process and consists of awareness, interest, evaluation, trial and finally adoption.
In general; customers will seek out, those financial products and suppliers
which offer the best value for money and they are educated about it. Earlier
studies suggest that customers adoption of electronic banking technologies
may be related to a number of factors, some associated with the characteristics
of the product or service and others associated with the characteristics of
the customers. For example, Gerrard and Cunningham (2003)
used 41 statements to measure the characteristics related to the adoption of
electronic banking and innovations by Singaporean customers. In another study
in Singapore, Liao and Cheung (2002) found that individual
expectations regarding accuracy, security, transaction speed, user-friendliness,
user involvement and convenience were the most important quality attributes.
Mattila et al. (2003) added that household income
and education had a significant effect on the adoption of online banking among
customers in Finland. Sathye (1999) identified security
concerns and lack of awareness as the main obstacles to non-adoption of Australian
Research variables: The expected relationship with regard to demographic,
personal, behavioral and attitudinal characteristics can be outlined in the
following variables. These variables will be examined in order to compare the
adopters and non-adopters beliefs on the adoption of IB services.
The variables were measured on the basis of the individuals own self-assessment
of how and why they perceive technology based financial services and their likelihood
to use them. The variables are as followed:
||Variables related to demographic characteristics are; age,
gender, marital status, educational level, occupation and household income
||Variables related to personal characteristics are; access to Internet,
aim of Internet usage, delivery channel preference, lack of awareness from
benefits, lack of technological skill, lack of security and trust
||Variables related to IB usefulness are; transferring money within the
same bank branches, transferring money to other banks, foreign exchanging
and buying/selling stock, utility payments and getting information inquiry
||Variables related to IB characteristics are; convenient, less cost, security
and privacy, ease of use, transaction speed, no need for carrying cash,
suitable for any type of transaction, satisfy all the banking and financial
MATERIALS AND METHODS
Questionnaire: The questionnaire includes two sections. The first section
gathers demographic information such as age, gender and etc. The second section
has three parts. Part one asks customers about their personal information for
their experience in using Internet and online banking. Part two asks the respondents
attitudes about the usefulness of banking services and part three focuses on
the respondents satisfaction level about the IB services. The questionnaires
were first pre-tested among 35 customers to comment on the relevance and wording
of the questionnaires items, length of survey and time taken to complete
the questions and it was also reviewed carefully by professionals with extensive
experience. The comments and suggestions of these professionals regarding the
clarity, validity and consistency of the questions are incorporated into the
survey instrument. Some of the questionnaire items were dropped, its layout
was modified and the wording of some of the questionnaire was changed to improve
the quality. Finally another test was carried out in one branch of the bank
among 30 customers. The Cronbachs alpha value was calculated for these
questionnaires and it was equal to 0.85 (over the recommended level of 0.70).
Sampling and data collection: Data collection was conducted from September to November 2006, in four different branches of Mellat bank in Tehran. The population of interest is defined as two groups of Mellat banks retail customers. Group one are those customers that have already adopted IB services (users) and group two are those who havent adopted IB services (non-users). The method of sampling was random sampling. Total 500 questionnaires were distributed and each of responses received was screened for error, incomplete and missing responses. After the screening process was carried out 31 considered as unusable and totally 469 useable questionnaires were collected. These 469 respondents were later split to users and non users of IB services. This was done to know the reason why all the non users have not been using it and to know the problems that users were facing.
Data analysis techniques: Data reduction is implemented using factor
analysis techniques and predicting the category of outcome for individual cases
is achieved using logistic regression analysis (Norusis, 2008).
Development of the adoption characteristics: Factor analysis with varimax rotation is used to establish the dimensions appropriateness to model the adoption of IB services. Dimensions with eigenvalues of 1 or above are retained and different numbers of factors are emerged. Because the proposed approach involves criterion variables with two categories, for classification of variables, binomial logistic regression is implemented. Logistic regression analysis allows one to predict a discrete outcome from a set of variables that may be dichotomous, discrete, or a mix of any type of these. In this statistical technique the predictors do not have to be normally distributed, linearly related, or of equal variance within each group. The goal of logistic regression is to correctly predict the category of outcome for individual cases using the most parsimonious model. Since, most of the independent variables of this study are categorical, logistic regression is implemented.
The first six attributes are concerned with straightforward factual information.
They are shown in Table 1, together with possible responses.
The percentage value and significance level for each attribute are also listed.
These are useful measures for depicting of tendency, towards the top or bottom
of the scale. For instance; a percentage value of 22.77 for the age attribute
means that there were more respondents from the middle age than the older and
younger age groups. It can also be deduced from the table that in user group
approximately 66% of respondents were male and 34% female. Also, concerning
their occupational background, 59.78% of users were governments employees.
Likewise, approximately 60% of the respondents were well educated. For more
clarity, frequency, percentage and significance difference values of respondents
on straightforward factual information is presented in Table 1.
The following five attributes are concerned with peoples agreement or
disagreement with a number of statements concerned with behavior and attitudes.
The majority of the respondents (53.02%) had access to Internet from office.
Comparison of users and non-users proportions in this category revealed a significant
difference between these two groups. Almost 65.83% of the respondents were using
Internet for gathering information and news. An interesting piece of information
was that, branch counter with 56.93% was the most preferred delivery channel
of the customers. This finding indicated that the customers preferred traditional
retail banking channel at present time. Concerning respondent awareness from
benefits, results revealed significant difference among the means of the data
and users were significantly more aware compared to non-user corresponding group.
One of the main obstacles in online banking services usage was the lack of technical
information specifically PC skills and a significant difference between users
and non-users groups were observed. The most important obstacle in IB usage
was the lack of trust to financial services security and a great deal of skepticism
was detected among the customers. In this context approximately 42% of respondents
were worried and a significant difference between users and non-users groups
was appeared and user respondents were more security concerned than non-users
||Straightforward factual information
|**Significant at the 1% level of probability; *Significant
at the 5% level of probability; ns: Not significant
For variables related to IB usefulness, the research variables i.e., information
inquiry, transferring money within the same bank branches, transferring money
to other banks, foreign exchanging and buying/selling stock and utility payment
were subjected to the principal component analysis (correlation matrix approach)
using orthogonal rotation, specifically varimax rotation. Also, the KMO and
Bartletts sphericity was used to test the sample appropriateness. The
variables, which had loadings of less than 0.5, were excluded and dimensions
with eigenvalues of 1 or above; were retained.
Table 2 exhibits the correlation matrix of all the variables
and it shows that all variables were correlated well and there is no need to
eliminate any variable. Also, the large values of Bartletts sphericity
test (569.7, df = 10, p = 0.000) and KMO statistics (0.83) indicated the appropriateness
of factor analysis i.e., the sample was adequate. Moreover, determinant was
used to test for multicollinearity and since it was 0.128 and was greater than
0.00001; multicollinearity was not a problem for the data. Similarly;
Cooper and Schindler (2003) argued that while the correlation coefficients
in matrix table are less than 0.80, the multicollinearity could be ignored.
Table 3 shows only 1 factor that was extracted from the analysis
along with its eigenvalues, the percentage of variance attributed to the factor
(61.806%) and also exhibits the cumulative variance of the factor. Since, only
one component was extracted; the solution could not be rotated. From the user
respondents point of view, the five attributes could be viewed as only
one and is meant that IB services had great influence on making the financial
and banking transactions quick with more compatibility and convenience.
For logistic regression analysis, first all 5 variables were entered into the general model and subsequently a backward stepwise method regression was employed. The backward stepwise model was same as the general model. However, due to reduction in multicollinearity, the significance of the variables increased. Of the 5 variables entered, information inquiry, balance statement and credit card statement, was found to be significant at 5% level. Table 4 presents the logistic regression analysis results with a significant effect on the adoption of IB along with all related statistics. Based on the ß values; it appears that two variables have a positive effect on the odds of a person belonging to the IB users, whereas three variables have a negative effect.
The general model correctly classified 68.3% of cases, which means it was a
good predictor in 68.3% of the cases in differentiating the users from non-users
of IB services. In the second stage, the factor that was found to be significant
in the general model was again significant in this stage. This factor was information
inquiry for balance statements and credit card statements. The logistic regression
analysis for both general and backward stepwise method regression models showed
that information inquiry as one of the Internet banking services was an important
factor in the adoption process rather than other banking services.
||Correlation matrix result
||Total variance explained
||Results of logistic regression analysis
||Correlation matrix result
||Total variance explained
The remaining 10 attributes are concerned with peoples agreement or disagreement
with a number of statements concerned with behavior and attitudes. Table
5 shows the correlation matrix of all the variables related to IB characteristics
and it presents that all variables correlated well and none of the correlation
coefficient is exceptionally large, therefore, there is no need to eliminate
any variable. Also, the large values of Bartletts sphericity test (1026,
df = 45, p = 0.000) and KMO (0.84) indicate the appropriateness of factor analysis
i.e., the sample is adequate. The determinant is used to test for multicollinearity
and since it is .024 and greater than 0.00001, multicollinearity is not a problem
for the data.
To identify the major respondent segments among IB users, the data of relevant attributes were subjected to factor analysis. Dimensions with eigenvalues of 1 or above were retained and as shown in Table 6, two factors emerged. The variables were as followed: convenient, cost, security and privacy, easy to use, speed transaction, no need for carrying cash, more suitable, satisfy transaction, easy usage of others and recommend to others. Table 6 exhibits all the factors extractable from the analysis along with their eigenvalues, the percentage of variance attributed to each factor and also shows the cumulative variance of the factors and the earlier factors. In the final part of this table, the eigenvalues of the factors after rotation are displayed and thus has the effect of optimizing the factor structure and one consequence for these data is that the relative importance of the two factors is equalized. Before rotation, factor 1 account for considerably more variance than the other (42.56% compared to 13.24%). However, after extraction, it accounts for only 34.9% of variance compared to 20.89%, respectively.
Table 7 shows the component matrix before rotation and all the loadings less than 0.6 are suppressed in the output so there are blank spaces for many of the loadings less than 0.6. It can be observed from this table that cost, security and privacy, easy to use, speed transaction, more suitable, easy usage of others and recommend to others load on factor 1 and only convenient load on factor 2.
In the rotated factor matrix of Table 8, component 1 has
high coefficients for security and privacy, ease of use, quick transaction,
suitable for any banking and financial transactions, satisfying all banking
and financial needs, easy usage of others and recommending to others. Therefore,
this component may be labeled as, availability and easy usage of financial and
banking transactions. Component 2 is highly related with convenience, low/no
cost and no need for carrying cash. Thus, component 2 may be labeled as, flexibility
of banking transactions.
||Rotated component matrix
||Results of logistic regression analysis
One could summarize the data by stating that customers
appear to seek two major kinds of benefits from satisfaction of IB services:
First; availability and easy usage of financial and banking transactions and
second; flexibility of banking transactions.
In subsequent data analysis, logistic regression analysis was carried out.
Table 9 exhibits the logit model of the factors with a significant
effect on the adoption of IB. Based on the ß values it appears that five
variables have a positive effect on the odds of a customer belonging to the
IB users, whereas other five variables have a negative effect. While, customers
who perceive more secure and convenience in Internet banking and its rather
straightforward use have significantly higher odds to actually adopt IB. In
performing logistic regression analysis, first all 10 variables were entered
into the general model and secondly a backward stepwise method regression was
carried out. Of the 10 variables entered, 4 were found to be significant at
5% level. The significant factors were convenience, no need for cash, usefulness
The general model correctly classified 71.2% of cases, which means it was a good predictor in 71% of the cases in differentiating the users from non-users. To simplify the model, the regression backward method was conducted with all 10 variables. All the factors that were found to be significant in the general model were again significant in this stage. The model correctly classified 71.5% of the cases. The logistic regression analysis for both general and backward stepwise method regression models showed that convenience, no need for carrying cash, usefulness and satisfaction were more important factors in the adoption process of IB services rather than other IB characteristics.
Several scholars have analyzed mainly the demographic characteristics of IB
customers but lesser emphasis has been placed on analyzing behavioral, attidunial
and social characteristics of the bank clients. The empirical findings of the
current study show that these characteristics have effective impacts on adoption
of IB services. In terms of personal and social characteristics, this study
contributes to this purpose, by identifying the Iranian customers attitude
toward IB services. Among demographic characteristics, more significant influence
is found for gender, educational level and monthly income. Gender has not been
found to have a direct effect on adoption of technology in general (Gefen
and Straub, 2000), whereas the results of current study reveals that males
represent the segment with the highest use of IB services. This finding is in
agreement with the results of Faith et al. (2007), who argue that gender
could affect the intention to adopt e-commerce including online banking. Besides,
higher use of IB has been evident for those who have higher educational level.
Also, a similar trend of IB usage is detected to those customers with middle-class
income. With notice to the large number of customers belonging to this class
of monthly income, if the banks manager strategy approaches towards these
customers, by providing special offers, it may be successful to help increase
the adoption rate of IB services. Therefore, bank managers should focus their
promotion activities aimed at such customers.
The majority of the customers are very comfortable and willing to use IB services.
Hence, it is very important for Iranian banks to have online banking services.
It is well-accepted fact that; providing good customer service and convincing
the customers about the non-economic relative advantages such as easy and safe
payment and protecting from theft or losing of cash, will increase the number
of adopters after a while. Based on the results, negative perceptions of the
Internet banking services such as risk of safety and risk of error have detrimental
impacts on adoption. So, bank managers should focus their promotional activities,
aimed to counteract such negative perceptions. The finding about the impact
of age on adoption of online banking services indicates that the effect of age
is not prominent. Therefore, age is not a crucial variable for banks that are
planning to offer IB services. In comparative terms; lack of trust to financial
services security was not rated differentially. Gefen and
Straub (2003) argue that trust is a crucial factor in many transaction relationships.
Generally, customers do not trust e-banking for some reasons especially due
to lack of the security of the system (Rotchanakitumnuai
and Speece, 2003). The results of current study revealed that all the customers
are very concerned about security in transaction processes. This result is in
consistent with the results, which have been reported earlier by other scholars
(Black et al., 2001; Lee and
Turban, 2001; Polatoglu and Ekin, 2001; Shah Alam
et al., 2007). Thus, in order to lure non-users respondents into
IB user group, bank manager needs to explain the high level of security, which
supports the banks computer system from third-party intrusions into an
Internet account. Another issue, which should be considered, is unequal level
of PC skill between users and non-users customers. PC proficiency has significant
influence on adoption. Hence, mass adoption and widespread diffusion of IB will
only occur when such technological barrier is lowered. People with higher PC
self-efficacy are more readily prepared to use online banking services. Since
there is a great number of Iranian who has no IB account at present, the potential
for IB in Iran is high. To achieve these outcomes, the study identifies a number
of strategies which bank managers could follow. While, this study is a maiden
attempt in this aspect in Iran; the result could be served as a baseline data
for further research and future researches can hopefully look for further avenues
in exploring the areas.
CONCLUSION AND LIMITATION
The majority of the customers are very comfortable and willing to use IB services. This is an important issue for Iranian banks to have high quality online banking services. Out of many factors that have been tested to be influencing the adoption of IB in Iran, four factors have been identified as more significant factors. The analysis shows that security concerns and lack of awareness stand out as the reasons for non-adoption of IB services by Iranian customers. These are the main reasons for not opening online banking or investment accounts. Thus, in order to lure non-users respondents into IB user group, bank manager needs to explain the high level of security, which supports the banks computer system from third-party intrusions into an Internet account. Among demographic variables; further significant influences have been found for gender and the type of occupation. In this context, the higher rate of using Internet banking services has been found for government employees. Regarding the large number of governmental organizations in Iran, if banks provide special offers for this segment, it may be successful to help increase the adoption rate of IB. This measure could help in rapid migration of customers to IB, resulting in considerable saving in operating cost for bank.
The key contribution of this study is the identification of prominent demographic, behavioral and attitudinal characteristics as predictor of customer choice in the context of IB services in Iran. This could ultimately lead to enhanced abilities for bank managers to develop more secure and trustable products. Also, this study sets an important benchmark for further research in the area.
Since, the sample size is limited; the findings can be taken only as indicative results and the findings have to be compared and confirmed with a research with much bigger sample size to obtain better accuracy. Some of the respondents do not know much about Internet banking and thereby rejected the questionnaire.
The authors are grateful to Tarbiat Modares University (TMU) and Luleå University of Technology (LTU) for financial support and research facilities. The grant No. is 403709.