Determining the relative importance of mobile banking quality factors

Computer Standards & Interfaces 35 (2013) 195–204

Contents lists available at SciVerse ScienceDirect

Computer Standards & Interfaces journal homepage: www.elsevier.com/locate/csi

Determining the relative importance of mobile banking quality factors Hsiu-Fen Lin ⁎ Department of Shipping and Transportation Management, National Taiwan Ocean University, No. 2, Beining Road, Keelung 202-24, Taiwan, ROC

a r t i c l e

i n f o

Article history: Received 7 May 2012 Received in revised form 1 July 2012 Accepted 26 July 2012 Available online 10 September 2012 Keywords: Mobile banking Quality evaluation Fuzzy analytic hierarchy process Extent analysis approach

a b s t r a c t The aim of this study is to use fuzzy analytic hierarchy process (AHP) with an extent analysis approach to develop a fuzzy evaluation model which prioritized the relative weights of m-banking quality factors between low- and high-experience groups. The research findings indicated that there are some similarities and differences between high- and low-experience groups with regard to the evaluation of m-banking quality. With respect to the final weights for the criteria level, both groups considered “customer service” to be the important factor affecting m-banking effectiveness. The research findings also provide insightful information to m-banking service providers so that they may improve the effectiveness and efficiency of m-banking. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Mobile banking (m-banking) (Internet banking using mobile devices, also known as m-banking, mbanking, SMS banking, etc.) can perform account balances and transaction history inquiries, funds transfers, and bill payments via mobile devices such as cell phones, smartphones, and PDAs (personal digital assistants) [24,49]. M-banking may have new features (such as ubiquity, flexibility and mobility) compared to conventional banking channels (e.g., automated teller machine, phone-banking, non-mobile Internet banking). For customers, m-banking provides a very convenient and effective means of managing personal finances, supporting seamless anytime, anywhere connectivity [40]. Since m-banking services are relatively new electronic delivery channels to be offered by banks, assessing their quality or effectiveness is critical for both m-banking researchers and service providers (including banks, telcos and other financial institutions). Chung and Kwon [13] and Lin [30] also argued that customer experience with m-banking services influence their intentions to use such services. To our best knowledge, no study has been undertaken to evaluate the relative importance of m-banking quality factors between two groups of customers — one group with low m-banking experience and the other with high m-banking experience. Information system (IS) researchers have proposed that m-banking can be considered as one of the most significant service innovations, which is emerging as a key platform for expanding access to banking

⁎ Department of Shipping and Transportation, Management, National Taiwan Ocean University, No.2, Beining Road, Keelung 202‐24, Taiwan. Tel.: +886 2 24622192x3409; fax: +886 2 24631903. E-mail address: hfl[email protected]. 0920-5489/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.csi.2012.07.003

transactions via mobile devices [25,47]. Thus, providing a professional and high quality service is increasingly recognized a critical factor for successful implementation of m-banking. Previous studies on m-banking quality have generally focused either on specific sets of technical functions or basic content management [26,56]. Regarding technical functions, there are evaluations of ease of use [29,31], transaction speed [16], and system security [53]. In analyzing mobile content delivery, some researchers have evaluated the up-to-date, accurate and relevant information [32,48]. Stair and Reynolds [45] also identified that determinants of m-banking quality or effectiveness are related to convenience, service accessibility without constraints of time and place, privacy and savings in time and effort. The literature has mentioned a broad range of factors that influence m-banking quality. The evaluation of m-banking quality is potentially very complex due to the multitude of variables that influence the decision process [55]. Determining the most important influences on m-banking quality is crucial and helps service providers focus on factors with the highest weight and identify the best policy to improve m-banking effectiveness. That is, how to evaluate the relative importance of these factors thus can be considered as a multiple-attribute decision‐making problem. Analytic hierarchy process (AHP) is an appropriate method for solving multiple-attribute decision‐making problems [41]. However, the decision maker can specify preferences in the form of natural language expressions about the importance of each evaluation item [17]. This implies that human judgment on the importance of alternatives or criteria is always subjective and imprecise. To make up for this deficiency in AHP, several researchers integrate fuzzy logic theory with AHP to determine the criteria weights from subjective judgments of decision makers [5,15,30,39]. Consequently, this study attempts to apply the fuzzy AHP approach to determine the relative weights of m-banking quality factors between low and high‐experience groups.

196

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

As a conclusion of the above statements, this study has two sub-objectives. The first is to evaluate the relative importance of m-banking quality factors. Based on reviewing and analyzing literature on m-banking quality factors, this study identifies evaluated criteria which influence the m-banking quality. Then this study integrates fuzzy AHP with an extent analysis approach to develop a fuzzy evaluation model which prioritized the relative weights of m-banking quality factors. By examining this gap, this study contributes to providing some design guidelines and strategies for firms involved in m-banking. 2. Literature review 2.1. Mobile banking M-banking is a subset of mobile commerce applications which enables customers to conduct conventional (e.g., balance checks and fund transfers) and more advanced (e.g., stock trading and portfolio management services) banking transactions through mobile devices [23,24,49]. M-banking adoption differs from electronic banking (or non-mobile Internet banking) adoption in at least two ways. First, electronic banking is Internet-based customer access to banking services, while m-banking is mobile phone based customer access to banking services. As electronic banking edges further into the mainstream of banking services, financial institutions are leading the way into the new technological frontier: mobile access [35]. M-banking has become increasingly ubiquitous and thereby changes the business of retail banks significantly in terms of cost reduction and increased convenience for the customer [33]. Second, the difference between mobile and electronic banking is the pace of evolution, with mobile banking evolving much faster than electronic banking [24]. M-banking services provide customer value creation due to being inherently time and place independent, as well as their effort-saving qualities [35]. Consequently, m-banking has become the self-service delivery channel that allows banks to provide information and offer services to their customers with more convenience via mobile devices. 2.2. Quality factors in the m-banking context According to DeLone and McLean's [14] IS success model, quality factors may be the important antecedents of IS success. Previous researchers have explored the quality factors associated with various mobile commerce applications. Mahatanankoon et al. [34] suggested that in order to reach mobile commerce benefits, operation modes and strategies must provide good service quality through valueadded, location-centric, and customized mobile applications. Chen [10] conducted a study on mobile payment and found quality factors (e.g., perceived transaction convenience, perceived transaction speed, security concerns, and privacy concerns) as important antecedents of customer acceptance of mobile payment. Choi et al. [12] found that transaction process and content reliability significantly influence customer satisfaction and loyalty in mobile commerce. In the study of Yeh and Li [54], they showed a close link between web site quality (i.e. interactivity and customization) and customer satisfaction towards the vendors on the mobile internet. Lu et al. [32] proposed and examined a multidimensional and hierarchical model of mobile service quality in the context of mobile brokerage services. Their measure of mobile service quality has three dimensions including interaction quality which is influenced by attitude, expertise, problem solving, and information, environment quality which is affected by equipment, design, and situation, and outcome quality which is determined by punctuality, tangibles, and valence. In the context of mobile information and entertainment services, Tan and Chou [48] proposed seven mobile service quality including perceived usefulness,

perceived ease of use, content, variety, feedback, experimentation, and personalization. The concept of m-banking quality can be defined as overall customer evaluations and judgments regarding the excellence and quality of mobile content delivery in the context of m-banking [14,28,44]. Several studies examined m-banking quality factors by studying the antecedents of customer satisfaction and behavioral intention to use m-banking. For example, Gu et al. [16] adapted the technology acceptance model (TAM) to study the determinants of customer intentions to use m-banking. In their research model, system quality was measured through perceived network speed and system stability of m-banking services. Lee and Chung [26] considered three quality factors (including system quality, information quality, and interface design quality) in which m-banking was provided and adapted from the updated DeLone and McLean IS model [14], to fit characteristics of m-banking services. Yu and Fang [55] also identified six dimensions to measure post-adoption customer perceptions of m-banking services including security service, interactivity, relative advantage, ease of use, interface creativity, and customer service, which were confirmed by exploratory and confirmatory factor analysis. Luo et al. [33] showed that both trust belief and risk belief significantly drive customer's intention to adopt m-banking. Zhou [57] used elaboration likelihood model (ELM) as the theoretical based to examine the effect of central cues (information quality and service quality) and peripheral cues (system quality, reputation, and structural assurance) on m-banking user behavior. These quite different perspectives regarding m-banking quality factors can roughly be termed content-based feature such as information quality versus functionally-based feature such as interface design and customer service. The existence of different influences on m-banking quality can provide a basis for assessing the relative importance of m-banking quality factors. Table 1 summarizes such factors discussed in the recent mobile commerce and m-banking literature. 3. Fuzzy analytic hierarchy process 3.1. Essences of fuzzy analytic hierarchy process Analytic hierarchy process (AHP) is a useful method for solving complex decision-making problems involving subjective judgment [43]. In AHP, the multi-attribute weight measurement is calculated via pairwise comparison of the relative importance of two factors. Though AHP is designed to capture decision-maker knowledge, the conventional AHP does not fully reflect human thinking style [4]. However, it is well recognized that human perceptions and judgments are represented by linguistic and imprecise patterns for a complex problem. Linguistic and imprecise descriptions were difficult to solve using AHP until the recent development in fuzzy decision-making [3,11]. Fuzzy set theory resembles human reasoning in its use of approximate information and uncertainty in decision generation. A major contribution of fuzzy set theory is its capability to represent vagueness. Meanwhile, AHP was developed to solve the multiple-attribute decision-making problem. By incorporating fuzzy set theory with AHP, fuzzy AHP enables a more accurate description of the multiple-attribute decision-making process [2]. The earliest work in fuzzy AHP appeared in van Laarhoven and Pedrycz [50], compared fuzzy ratios described with triangular membership functions. Many studies using fuzzy AHP are proposed to calculate the importance (weights) of evaluation items [5,15,30,39]. Therefore, in this study, the author prefers the fuzzy AHP approach since this approach is adequate to explicitly capture the importance assessment for human imprecise judgments. Since mobile communication technologies and wireless transaction environments involve intangibility and uncertainty, evaluating m-banking quality becomes more difficult for customers. Most decision makers tend to assess performance based on their own

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

197

Table 1 M-banking quality factors. M-banking quality factors

References Chen [10]

Choi et al. [12]

Gu et al. [16]

Response time Mobility Security Reliability Responsiveness Completeness Trust Multimedia capability Accessibility Accuracy Currency Relevance Understandability Navigability Empathy Format

● ● ●

●

●

● ●

Kar et al. [20]

● ● ● ● ●

Kim et al. [23]

Lee and Chung [26]

● ● ●

● ●

●

●

●

Lu et al. [32]

Tan and Chou [48]

●

Vlachos and Vrechopoulos [51]

Yeh and Li Yu and [54] Fang [55]

● ● ●

●

●

● ●

● ● ● ● ● ●

● ● ●

●

subjective and imprecise judgments [2]. Previous studies apply fuzzy set theory to deal with quality measurement [52,57] and website design [27]. Furthermore, Marler et al. [36] examined differences between the determinants of initial adoption and subsequent continued usage of employee self-service technology. Kim and Oh [22] argued that the testing model for the use and impact of mobile data services across different customer experiences (e.g., non-adopters versus adopters) might result in new insights. As a result, understanding of the relative importance of m-banking quality factors across different experience groups is necessary. For these reasons, this study applied a fuzzy AHP approach to establish a fuzzy evaluation model which determines the relative importance of m-banking quality factors between low- and high-experience groups. 3.2. Constructing aggregate fuzzy judgment matrix The design of the questionnaire incorporates pair-wise comparisons of decision elements within the hierarchical structure. Each evaluator is asked to express relative importance of two criteria in the same level by a nine-point rating scale. Collect the scores of pairwise comparison, and form pair-wise comparison matrices for each of the H evaluators. The scores of pair-wise comparison are transformed into linguistic variables, which are represented by triangular fuzzy numbers (see Table 2) [3]. The arithmetic mean value is applied to integrate fuzzy

Table 2 Triangular fuzzy numbers. Linguistic variables

Triangular fuzzy numbers

Equally important Intermediate Weakly more important Intermediate Strongly more important Intermediate Very strongly more important Intermediate Absolutely more important

(1,1,1) (1,2,3) (2,3,4) (3,4,5) (4,5,6) (5,6,7) (6,7,8) (7,8,9) (9,9,9)

●

● ● ● ●

● ●

●

●

●

● ● ●

●

● ● ● ● ● ●

●

Zhou [56]

● ● ●

● ●

● ●

values of H evaluators. An aggregate fuzzy judgment matrix can be defined as: h i M ¼ M ij

ð1Þ

where M an aggregate fuzzy judgment matrix of H evaluators, Mij the combining fuzzy assessments between criterion i and criterion j of H evaluators,

  Mij ¼ lij ; mij ; uij ;

n number of the related criteria at the this level, Mij = (1,1,1), ∀ i = j; and Mji = 1 M , ∀ i, j = 1, 2, …, n ij

3.3. Extent analysis method on fuzzy AHP The extent analysis method is used to consider the extent of an object to be satisfied for the goal, this is, satisfied extent. In the method “extent” is quantified by using fuzzy number. On the basis of the fuzzy values for the extent analysis of each object, a fuzzy synthetic degree value can be obtained, which is defined as follow: Let X = {x1,xx, …,xn} be an object set, and U = {u1,ux, …,um} be a goal set. According to the method of Chang's [7,8] extent analysis model, each object is taken and extent and extent analysis for each goal, gi, is performed respectively. Therefore, m extent analysis values for each object can be obtained and shown as follows: 1

2

m

Mgi ; M gi ; ⋯; M gi ; i ¼ 1; 2; …; n

ð2Þ

where all the Mgi j(j = 1, 2, …, m) are triangular fuzzy numbers. The steps of the improved Chang's extent analysis model [58], which is applied in this study, can be given in the following steps:

198

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

1. The value of fuzzy synthetic extent with respect to the ith objects is defined as: Si ¼

m X

2 n X m X

j Mgi ⊗4

j¼1

3−1

T

j Mgi 5

ð3Þ

i¼1 j¼1

To obtain

m X

j M gi ,

perform the fuzzy addition operation of m extent

analysis values for a particular matrix such that: 0 1 m m m X X X j @ Mgi ¼ lj ; mj ; uj A

j¼1

j¼1

j¼1

ð4Þ

j¼1

2 3−1 n X m X j and to obtain 4 M gi 5 , perform the fuzzy addition operai¼1 j¼1

tion of Mgi j(j = 1, 2, …, m) values such as: n X m X

j

Mgi ¼

i¼1 j¼1

n X

li ;

i¼1

n X i¼1

W ¼ ðdðA1 Þ; dðA2 Þ; ⋯; dðAn ÞÞ

mi ;

n X

! ð5Þ

ui

i¼1

4. Research model To validate the main influences on m-banking quality, measurement items were developed using the research literature dealing with these factors (see Table 3). A questionnaire was conducted to verify the factors that had been identified in the literature, with the aim of investigating their degree of importance. Data for this study was collected by the means of a survey conducted in Taiwan. The paper-based questionnaires were distributed to 300 m-banking customers provided by one public and three private banks. Respondents were asked to indicate the extent to which they felt various factors influenced their evaluations Table 3 Results of exploratory factor analysis. Measurement items

and then compute the inverse of the vector in Eq. (5) such that: 0 2 3−1 B n X m X B 1 j 4 Mgi 5 ¼ B BX n @ i¼1 j¼1 i¼1

ð12Þ

where W is not a fuzzy number.

j¼1

m X

4. After normalization, the normalized weight vectors are obtained as follows:

1 C 1 1 C ; n ; n C X X C A ui mi li i¼1

ð6Þ

i¼1

2. The degree of possibility of M2 = (l2,m2,u2) ≥ M1 = (l1,m1,u1) is defined as: h

i V ðM2 ≥M 1 Þ ¼ sup uM1 ðxÞ; uM2 ðyÞ y≥x

ð7Þ

and can be equivalently expressed as follows: V ðM2 ≥M 1 Þ ¼ hgt ðM1 ∩M 2 Þ ¼ uM2 ðdÞ 8 1; if m2 ≥m1 > > < 0; if l1 ≥u2 ¼ l1 −u2 > > ; otherwise : ðm2 −u2 Þ−ðm1 −l1 Þ

ð8Þ

where d is the crossover point's abscissa for M1 and M2. To compare M1 and M2, we need both the values of V(M1 ≥ M2) and V(M2 ≥ M1).

3. The degree of possibility for a convex fuzzy number to be greater than k convex fuzzy numbers Mi(i = 1, 2, …, k) can be defined by: V ðM≥M1 ; M 2 ; …; Mk Þ ¼ V½ðM≥M1 Þ and ðM≥M2 Þ and …andðM≥Mk Þ ¼ minV ðM≥Mi Þ; i ¼ 1; 2; …k

ð9Þ

Assume that: ′

d ðAi Þ ¼ minV ðSi ≥Sk Þ

ð10Þ

for k = 1, 2, …, n; k ≠ i. Then the weight vector is obtained as follows:  T ′ ′ ′ ′ W ¼ d ðA1 Þ; d ðA2 Þ; ⋯; d ðAn Þ where Ai(i = 1, 2, …, n) are n elements.

ð11Þ

Functionality (Cronbach's α value = 0.954) Accessibility: m-banking makes banking transactions easy to access. Response time: the waiting time for loading m-banking transactions is reasonable. Mobility: m-banking is accessible at anytime and anywhere. Security: m-banking provides enough security to conduct banking transactions. Content (Cronbach's α value = 0.887) Accuracy: information provided by m-banking is accurate. Currency: information provided by m-banking is always up to date. Relevance: m-banking provides relevant information about banking transactions. Completeness: m-banking provides customers with a complete set of information. Customer service (Cronbach's α value = 0.850) Reliability: m-banking provides the right solution to customer requests. Responsiveness: m-banking is respon sive to customer inquiries. Trust: using mobile devices in banking transactions is trustworthy. Empathy: according to customer transaction history, m-banking provides individual attention to customers. Interface design (Cronbach's α value = 0.799) Multimedia capability: m-banking provides an appropriate multimedia (such as graphic and image) presentation. Format: The contents of m-banking transactions (such as range, depth and structure) are clearly presented on the screen. Understandability: the presentation style of m-banking is easy to understand. Navigability: m-banking has an easy navigation to find information.

Factor loading

Eigenvalue Cumulative percentage of variance 3.295

20.596%

2.812

38.171%

2.386

53.083%

2.292

67.408%

0.747 0.831

0.846 0.859

0.759 0.839 0.647

0.823

0.508 0.536 0.818 0.763

0.554

0.842

0.761

0.578

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

of m-banking quality, using a seven-point Likert scale (ranging from 1 = strongly disagree to 7 = strongly agree). Of the 300 questionnaires distributed, 190 completed and usable questionnaires were returned, representing a response rate of 63.3%. Of the 190 usable respondents, 25% were less than 30 years of age (n = 48), 54% were 31 to 40 (n = 102), and 21% were over 41 years old (n= 40). More than 80% respondents had used the m-banking for more than 1 year. Finally, about 48% of the respondents were male (n= 92) and 52% were female (n= 98). A survey instrument should be validated employing statistical techniques such as a reliability test in order to confirm the internal consistency of measures and factor analysis in order to confirm the construct validity, including both convergent and discriminant validity [46]. Cronbach α values were found to be ranging from 0.799 to 0.954 for all dimensions that exceeded Nunnally's [38] criterion of 0.7 (see Table 3). Hence, the scales for all constructs were deemed to exhibit adequate reliability. Based on application of exploratory factor analysis (EFA), the principal component factor analysis adopted for analysis and result undergoes varimax rotation to extract major factor dimensions. As shown in Table 3, the factor analysis results satisfied the attribute of construct validity including both the convergent validity (eigenvalues greater than 1, all item loadings greater than 0.5) and discriminant validity (cross loading of items less than 0.5). Thus, we conclude that the scales should have sufficient construct reliability and validity. Moreover, after the axis is rotated, four attributes, “accessibility”, “response time”, “mobility”, and “security”, have a higher loading factor in the first dimension that we rename “functionality”. Four attributes “accuracy”,

199

“currency”, “relevance”, and “completeness” form a new dimension “content”. Attributes “reliability”, “responsiveness”, “trust”, and “empathy” are integrated as a new dimension “customer service”. The “multimedia capability”, “format”, “understandability”, and “navigability” are then combined as “interface design”. Using confirmatory factor analysis (CFA), all 16 items had significant factor loadings (t-statistics at the 0.05 level) on their corresponding factors (see Fig. 1) [19]. The CFA was evaluated using the model-fit indices that provide empirical evidence of the degree of correspondence between the proposed model and the data [21]. For a measurement model to have sufficiently good model fit, the chi-square value normalized by degrees of freedom (χ 2/df) should not exceed 3 [1], Good Fit Index (GFI), Non-Normed Fit Index (NNFI) and Comparative Fit Index (CFI) should exceed 0.9, and the Root Mean Square Error of Approximation (RMSEA) should not exceed 0.10 [9]. The results of current CFA model are reported in Fig. 1, χ 2/df was 2.32 (χ 2 = 863.907; df = 372), GFI was 0.907, NNFI was 0.953, CFI was 0.962, and RMSEA was 0.071, suggesting adequate model fit. In order to apply the fuzzy AHP to the prioritizing of m-banking quality, all factors influencing m-banking quality are initially structured into different hierarchical levels. Thus, based on the analysis results presented above, this study developed a research model of analytic hierarchy (see Fig. 2). The goal is to evaluate the relative weights of the m-banking quality (Level 1). Level 2 indicates the four criteria which influence the m-banking quality. Finally, four criteria were created to assess the sixteen sub-criteria (Level 3).

Fig. 1. Results of confirmatory factor analysis.

200

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

Level 1: Goal

Level 2: Criteria

Level 3: Sub-criteria C11

C1

C12 C13

Functionality

C14 C21 C2

C22 C23

Content

C24

M-banking quality

C31 C3

C32 Customer service

C33 C34 C41

C4

C42 Interface design

C43 C44

Accessibility Response time Mobility Security Accuracy Currency Relevance Completeness Reliability Responsiveness Trust Empathy Multimedia capability Format Understandability Navigability

Fig. 2. Research model of analytic hierarchy.

5. Empirical analysis 5.1. Scaling the relative importance of the criteria An AHP questionnaire is designed in the form of a pair-wise comparison based on the research model of analytic hierarchy (see Fig. 2). A conventional AHP questionnaire format (nine-point rating scale) indicates the relative importance of each criterion (or sub-criterion) in the same hierarchy (see Appendix A for an example of the AHP questionnaire). Data for this study were collected using an AHP questionnaire survey administered to subjects recruited from universities, institutes, and companies. Furthermore, to focus the AHP process it has been advised to engage a small group of participants [42], as little as 10 participants are sufficient [37]. The experiment involved 27 evaluators who had used at least one m-banking service. All evaluators using a paper-based survey to evaluate the hierarchy model of the m-banking quality. After eliminating six incomplete replies, 21 usable questionnaires were received. The gender ratio was 38% male and 62% female, and the average of m-banking experience is 1.2 years. Evaluators were split into low- (defined as using m-banking a few times a month or less) and high- (defined as using m-banking a few times a week or more) experience groups. Of the 21 evaluators, Table 4 Aggregate fuzzy judgment matrix with respect to goal for the low-experience group. Goal C1 C1 C2 C3 C4

(1.000, 1.000) (0.372, 0.558) (0.957, 2.287) (1.149, 1.208)

1.000, 0.446, 1.543, 1.187,

C2

C3

(1.794, 2.241, 2.689) (1.000, 1.000, 1.000) (0.818, 0.947, 1.080) (0.483, 0.718, 1.395)

(0.437, 1.044) (0.926, 1.222) (1.000, 1.000) (0.367, 1.091)

C4 0.648, 1.056, 1.000, 0.551,

λ max = 4.2275; CI = 0.0758; RI(4) = 0.90; CR = 0.0842 ≤0.1

(0.828, 0.842, 0.870) (0.717, 1.393, 2.072) (0.917, 1.814, 2.722) (1.000, 1.000, 1.000)

nine were assigned to the low-experience group, whereas twelve were assigned to the high-experience group. The aim of the survey was to collect evaluator opinions to measure the relative weight among the influences on m-banking quality. Therefore, these evaluators were asked to fill the questionnaire up, and then analyzing their subjective judgments for m-banking quality factors. 5.2. Constructing the fuzzy AHP approach for the m-banking quality 5.2.1. Step1: constructing aggregate fuzzy judgment matrix After nine evaluators of the low-experience group finished their assessments of relative importance for the criteria of m-banking quality, the aggregate fuzzy judgment matrix for criteria with respect to goal is generated by Eq. (1), as shown in Table 4. 5.2.2. Step2: consistency test The result of the consistency test, the consistency ratio (CR) of the aggregate matrix of the nine evaluators is less than 0.1 (as shown in Table 4). Thus, the consistency of the aggregate fuzzy judgment matrix is acceptable. 5.2.3. Step3: calculating weight vectors for individual levels After verifying the consistency, this study adopted the extent analysis method [7,8] to calculate weight vectors for individual levels. 1. Eqs. (3)–(6) are adopted to obtain the values of fuzzy synthetic extents with respect to the goal, as shown below: SC 1 ¼ ð4:059; 4:731; 5:603Þ⊗ð1=22:238; 1=17:387; 1=13:765Þ ¼ ð0:183; 0:272; 0:407Þ SC 2 ¼ ð3:014; 3:894; 4:852Þ⊗ð1=22:238; 1=17:387; 1=13:765Þ ¼ ð0:136; 0:224; 0:352Þ SC 3 ¼ ð3:692; 5:304; 7:089Þ⊗ð1=22:238; 1=17:387; 1=13:765Þ ¼ ð0:166; 0:305; 0:515Þ SC 4 ¼ ð2:999; 3:457; 4:694Þ⊗ð1=22:238; 1=17:387; 1=13:765Þ ¼ ð0:135; 0:199; 0:341Þ

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

2. According to Eq. (8), the degrees of possibility are calculated as below:       V SC 1 ≥SC 2 ¼ 1; V Sc1 ≥Sc3 ¼ 0:879; V Sc1 ≥Sc4 ¼ 1       V SC 2 ≥SC 1 ¼ 0:779; V Sc2 ≥Sc3 ¼ 0:697; V Sc2 ≥Sc4 ¼ 1       V SC 3 ≥SC 1 ¼ 1; V Sc3 ≥Sc2 ¼ 1; V Sc3 ≥Sc4 ¼ 1       V SC 4 ≥SC 1 ¼ 0:684; V Sc4 ≥Sc2 ¼ 0:891; V Sc4 ≥Sc5 ¼ 0:623

Table 5 Weighted for m-banking quality factors (low-experience group/high-experience group). Weights of criteria levela

Local weights of sub-criteria level

Global weights of sub-criteria levelb

Functionality 0.275/ 0.255

Accessibility

0.071(5)/0.072(6)

3. For each pair-wise comparison, the minimum of the degrees of possibility if sound as below (see Eqs. (9) and (10)):   ′ d ðC 1 Þ ¼ minV SC 1 ≥SC 2 ; SC 3 ; SC 4 ¼ 0:879   d′ ðC 2 Þ ¼ minV SC 2 ≥SC 1 ; SC 3 ; SC 4 ¼ 0:697   d′ ðC 3 Þ ¼ minV SC 3 ≥SC 1 ; SC 2 ; SC 4 ¼ 1   ′ d ðC 4 Þ ¼ minV SC 4 ≥SC 1 ; SC 2 ; SC 3 ¼ 0:623 Thus, the weight vector is calculated as:

Content

Customer service

0.218/ 0.298

0.312/ 0.276

 ′ ′ ′ ′ T T W ¼ d ðC 1 Þ; d ðC 2 Þ; d ðC 3 Þ; d ðC 4 Þ ¼ ð0:879; 0:697; 1; 0:623Þ ′

4. Via normalization, the relative importance weights of the criteria with respect to goal for low-experience group are calculated as follows:  T T W ¼ dðC 1 Þ; dðC 2 Þ; dðC 3 Þ; dðC 4 Þ ¼ ð0:275; 0:218; 0:312; 0:195Þ

5.2.4. Step4: estimating weight vectors for sub-criteria Steps 1–3 are performed to calculate weight vectors for sub-criteria with respect to functionality (C1), content (C2), customer service (C3), and interface design (C4). 5.2.5. Step5: calculating overall criteria weights and obtaining final ranking In order to compare all m-banking quality factors at the same level of the hierarchical structure, the priority weights (including local weights and global weights) and ranking are also completed. The final priority weights and ranking of m-banking quality factors for low- and high-experience groups are summarized in Table 5. 6. Research findings and discussion This study utilized fuzzy AHP to examine the similarities and differences between high- and low-experience groups in terms of the evaluation of m-banking quality. With respect to the final weights for the criteria level as shown in Table 5, both high- and lowexperience groups considered “customer service” to be the important factor for evaluating m-banking quality (weight for low-experience group = 0.312, weight for high-experience group = 0.276). One of the biggest advantages of m-banking is convenience, being able to conduct banking transactions at anytime from anywhere. Both the low and high groups (i.e., experience levels) always expect to obtain ubiquitous and convenient m-banking services, the provision of service to customers before, during and after a banking transaction may become primary considerations. Customers are more likely to continue using m-banking when they obtain reliable, prompt delivery, trust worthy and personalized services. This result provides further support that the level of service quality has a significant effect on behavioral intention to use mobile internet services for customers [51]. The analytical results also indicated some differences in the evaluation of m-banking quality between high- and low-experience groups. For example, less-experienced customers weighted “functionality”

201

Interface design

0.195/ 0.171

0.257(2)/ 0.281(2) Response time 0.220(3)/ 0.224(3) Mobility 0.202(4)/ 0.201(4) Security 0.321(1)/ 0.294(1) Accuracy 0.319(1)/ 0.281(2) Currency 0.233(3)/ 0.278(1) Relevance 0.245(2)/ 0.246(3) Completeness 0.203(4)/ 0.195(4) Reliability 0.275(2)/ 0.226(4) Responsiveness 0.198(4)/ 0.295(1) Trust 0.287(1)/ 0.242(2) Empathy 0.240(3)/ 0.237(3) Multimedia 0.201(4)/ capability 0.232(3) Format 0.281(2)/ 0.253(2) Understandability 0.211(3)/ 0.209(4) Navigability 0.307(1)/ 0.306(1)

0.061(8)/0.057(11) 0.055(10)/0.051(13) 0.088(2)/0.075(4) 0.070(6)/0.084(1) 0.051(13)/0.083(2) 0.053(12)/0.073(5) 0.044(14)/0.058(10) 0.086(3)/0.062(9) 0.062(7)/0.081(3) 0.089(1)/0.067(7) 0.075(4)/0.066(8) 0.039(16)/0.040(15) 0.055(11)/0.043(14) 0.041(15)/0.036(16) 0.060(9)/0.052(12)

Notes: aLocal weight is derived from judgment with respect to a single criterion. b Global weight is derived from multiplication by the weight of the criteria. Ranking results are listed in parentheses.

(0.275) more important than “content” (0.218). By reviewing the global weights of the 16 sub-criteria in Table 5, “trust (0.089)”, “security (0.088)”, and “reliability (0.086)” are the top three rankings. However, for customers with high m-banking experience, the weight of “content” (0.298) was greater than that of “functionality” (0.255). By looking at the global weights of sub-criteria level for the high-experience group, “accuracy (0.084)”, “currency (0.083)”, and “responsiveness (0.081)” are the top three rankings. This difference can be explained by the fact that both groups have different perceptions regarding the important determinants of m-banking quality. The less-experienced customers stressed the functionality attributes associated with m-banking services. This is consistent with extant research, which has reported the effect of functionality attributes on customer intentions to use m-banking [16,56]. Providing an operating environment with enhanced speed, ease-of-use and security are essential to building high-quality m-banking and reducing the risk of alienating customers with limited familiarity with m-banking. If m-banking has rich functionality, customers may believe service providers' ability and integrity to provide quality services. In contrast, customers who have more m-banking experience consider content quality to be the critical factor in facilitating m-banking success; this is consistent with other mobile commerce studies [6,26,51]. By providing appropriate information (including accurate and up-to-date financial information) and prompt customer service, service providers can increase the usefulness of m-banking, as well as the loyalty of experienced customers. Specially, since high-experienced customers usually navigate and access information concerning the latest transactions first, providing high quality information can facilitate the banking transaction process by helping them scan, filter, collect, and integrate financial resources.

202

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

In short, depending on the differences among learners with varying levels of cumulative m-banking experience, high-experience learners emphasized the factors used to evaluate website functionality. However, the factors used to evaluate website attractiveness received a higher weighting in less experienced online learners. Therefore, m-banking service providers must emphasize the balance between high-quality information and greater level of security across low‐ and high‐experience groups.

content quality. Improved content quality can facilitate m-banking transactions by enhancing the fit between financial information provided and customer information requirements. To retain existing customers, m-banking service providers should actively maintain and improve content quality to satisfy customer information needs, and mobile content delivery should accurately and promptly present financial information, while also facilitating customer financial management.

7. Conclusion

7.3. Limitations and future research

7.1. Theoretical contributions

There are several limitations to this study, requiring further examination and additional research. First, the evaluation criteria of m-banking quality were generated from the literature review and factor analysis, these methods may exclude some possible influences on m-banking effectiveness. Future research can use different methodologies, such as longitudinal studies, focus groups and interviews to identify other criteria of m-banking quality. Second, prior research has suggested a significant relationship between individual differences and Internet usage [18]. Future research can examine whether there are some differences between low- and high-revenue generating groups with regard to the evaluation of m-banking quality. Third, this study focuses on Taiwan, which is a small island economy compared with many others, some with far more banking institutions. Future research can examine the m-banking quality evaluation model for different countries, thus proves the practicality of the fuzzy evaluation procedure proposed by this study. Finally, this study uses the fuzzy AHP with an extent analysis approach to develop an evaluation model which helps service providers understand the critical factors in implementing successful m-banking. Future research could examine moderation test with either structural equation modeling or regression techniques to make comparisons of different groups and compare the results with the fuzzy AHP analysis.

Although the literature has contributed to identifying many quality factors influencing usage intention of m-banking, little is known regarding their relative importance across different m-banking experience groups. This study attempts to bridge this gap by providing a fuzzy evaluation model which prioritized the relative weights of m-banking quality factors between low- and high-experience groups. Major contributions are summarized below. First, the evaluation of m-banking quality can be considered as a complex multi-attribute decision-making problem. This study conducts the literature review and factor analysis to generate a hierarchical structure for m-banking quality, which includes 16 subcriteria along with four criteria (e.g., functionality, content, customer service, and interface design). Second, the fuzzy AHP with an extent analysis approach is proposed to determine the relative importance of m-banking quality factors across different customer experiences. The research findings of this study indicated that there are some differences between low- and high-experience groups with regard to the evaluation of m-banking quality. The evaluation factors of customer service received a higher weighting in less-experienced customers, while high-experience customers emphasized the evaluation factors of content quality. Analysis of the evaluation results can provide guidance to service providers in identifying the key factors facilitating m-banking development and find the best policy for improving system effectiveness. 7.2. Practical implications First, low- and high-experience customers generally consider service quality essential to a high quality m-banking service, thus providing quick, reliable, convenient, and personalized services can increase customer trust in m-banking. In terms of less experienced customers, they may feel apprehensive about using m-banking, for reasons that include lack of encryption of SMS (short message services) messages and customer fear of disclosing personal data. Reducing this fear, as well as the potential risks associated with wireless transaction, requires increasing trust, security, and service reliability to retain less experienced customers. Service providers thus must reduce customer fears associated with m-banking by providing financial loss protection policy, customers with a satisfaction guarantee, and personalized information and attention. As for highly experienced customers, they expect high responsiveness to their concerns and inquiries. Thus, m-banking service providers should offer prompt and courteous customer service, to cater to the needs of highly experienced customers. Second, to increase less experienced customer perceptions of m-banking functionality, it is recommended that advertising materials include customer interview scripts and reviews or survey reports. M-banking service providers should also pay attention to practical solutions and provide advanced functionalities (e.g., intelligent agent-based portfolio management and financial planning services) to enhance competence beliefs regarding service providers, subsequently reducing the risk of losing customers. Third, catering to highly experienced customers interested in using m-banking, it is necessary to strongly emphasize the importance of

Appendix A. An example of question items in AHP questionnaire Please compare in pairs the relative importance between two given item statements regarding the mobile banking (m-banking) quality. If a criterion (or sub-criterion) on the left is more important than the one matching on the right, put your check mark to the left of the importance “Equal” under the importance level you prefer. If a criterion (or sub-criterion) on the left is less important than the one matching on the right, put your check mark to the right of the importance “Equal” under the importance level you prefer. The notations of relative importance are following: (1) (2) (3) (4) (5)

Absolutely-Absolutely more important Very Strongly-Very strongly more important Strongly-Strongly more important Weakly-Weakly more important Equally-Equally important

With respect to the overall goal “evaluation of m-banking quality“ Q1: How important is functionality (C1) when it is compared with content (C2)? Q2: How important is functionality (C1) when it is compared with customer service (C3)? Q3: How important is functionality (C1) when it is compared with interface design (C4)? Q4: How important is content (C2) when it is compared with customer service (C3)? Q5: How important is content (C2) when it is compared with interface design (C4)? Q6: How important is customer service (C3) when it is compared with interface design (C4)?

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204 With respect to: m-banking quality

Importance of one criterion over another

Questions

Criteria

Absolutely

Very strongly

Strongly

Weakly

9:1

7:1

5:1

3:1

Q1 Q2 Q3 Q4 Q5 Q6

C1

8:1

6:1

4:1

2:1

C2 C3

References [1] R.P. Bagozzi, Y. Yi, On the evaluation of structural equation model, Journal of the Academy of Marketing Science 16 (1) (1988) 74–94. [2] F.T. Bozbura, A. Beskese, C. Kahraman, Prioritization of human capital measurement indicators using fuzzy AHP, Expert Systems with Applications 32 (4) (2007) 1100–1112. [3] J.J. Buckley, Fuzzy hierarchical analysis, Fuzzy Sets and Systems 17 (3) (1985) 233–247. [4] G. Buyukozkan, Multi-criteria decision making for e-marketplace selection, Internet Research 14 (2) (2004) 139–154. [5] G. Buyukozkan, Determining the mobile commerce user requirements using an analytic approach, Computer Standards & Interfaces 31 (1) (2009) 144–152. [6] M. Chae, J. Kim, H. Kim, H. Ryu, Information quality for wireless internet services, Electronic Markets 12 (1) (2002) 38–46. [7] D.Y. Chang, Extent analysis and synthetic decision, in: Optimization Techniques and Applications, 1, World Scientific, Singapore, 1992, p. 352. [8] D.Y. Chang, Applications of the extent analysis method on fuzzy AHP, European Journal of Operational Research 95 (3) (1996) 649–655. [9] P.Y.K. Chau, P.J.E. Hu, Information technology acceptance by individual professionals: a model comparison approach, Decision Sciences 32 (4) (2001) 699–719. [10] L.D. Chen, A model of consumer acceptance of mobile payment, International Journal of Mobil Communications 6 (1) (2008) 32–52. [11] C.H. Cheng, Evaluating weapon systems using ranking fuzzy number, Fuzzy Sets and Systems 107 (1) (1999) 25–35. [12] J. Choi, H. Seol, S. Lee, H. Cho, Y. Park, Customer satisfaction factors of mobile commerce in Korea, Internet Research 18 (3) (2008) 313–335. [13] N. Chung, S.J. Kwon, The effects of customers' mobile experience and technical support on the intention to use mobile banking, Cyberpsychology & Behavior 12 (5) (2009) 539–543. [14] W.H. DeLone, E.R. McLean, The DeLone and McLean model of information systems success: a ten-year update, Journal of Management Information Systems 9 (4) (2003) 9–30. [15] H.P. Fu, Y.C. Ho, R.C.Y. Chen, T.H. Chang, P.H. Chien, Factors affecting the adoption of electronic marketplaces: a fuzzy AHP analysis, International Journal of Operations & Production Management 26 (12) (2006) 1301–1324. [16] J.C. Gu, S.C. Lee, Y.H. Suh, Determinants of behavioral intention to mobile banking, Expert Systems with Applications 36 (7) (2009) 11605–11616. [17] J.F. Hair Jr., W.C. Black, B.J. Babin, R.L. Anderson, W.C. Tatham, Multivariate Data Analysis, Macmillan, New York, 2006. [18] W. Hong, J.Y.L. Thong, W.M. Wong, K.Y. Tam, Determinants of user acceptance of digital libraries: an empirical examination of individual differences and system characteristics, Journal of Management Information Systems 19 (3) (2001) 97–124. [19] C. Kahraman, U. Cebeci, Z. Ulukan, Multi-criteria supplier selection using fuzzy AHP, Logistics Information Management 16 (6) (2003) 382–394. [20] E. Kar, S. Muniafu, Y. Wang, Mobile services used in unstable environment: design requirements based on three case studies, in: Proceedings of the Eighth International Conference on Electronic Commerce, Fredericton, Canada, 2006. [21] M. Keil, B.C.Y. Tan, K.K. Wei, A cross-cultural study on escalation of commitment behavior in software projects, MIS Quarterly 24 (2) (2000) 299–325. [22] B. Kim, J. Oh, The difference of determinants of acceptance and continuance of mobile data services: a value perspective, Expert Systems with Applications 38 (3) (2011) 1798–1804. [23] G. Kim, B. Shin, G.G. Lee, Understanding dynamics between initial trust and usage intentions of mobile banking, Information Systems Journal 19 (3) (2009) 283–311. [24] T. Laukkanen, Internet vs mobile banking comparing customer value perceptions, Business Process Management Journal 13 (6) (2007) 788–797. [25] T. Laukkanen, Measuring mobile banking customers' channel attribute preferences in service consumption, International Journal of Mobile Communications 5 (2) (2007) 123–138. [26] K.C. Lee, N. Chung, Understanding factors affecting trust in and satisfaction with mobile banking in Korea: a modified DeLone and McLean's model perspective, Interacting with Computers 21 (5) (2009) 385–392. [27] Y. Lee, K.A. Kozar, Investigating the effect of website quality on e-business success: an analytic hierarchy process (AHP) approach, Decision Support Systems 42 (3) (2006) 1383–1401.

Equally

Weakly

1:1

1:2

1:3

203

Strongly

Very strongly

Absolutely

1:4

1:6

1:8

1:5

1:7

Criteria

1:9 C2 C3 C4 C3 C4 C4

[28] G.G. Lee, H.F. Lin, Customer perceptions of e-service quality in online shopping, International Journal of Retail and Distribution Management 33 (2) (2005) 161–176. [29] M.S.Y. Lee, P.F. McGoldrick, K.A. Keeling, F. Doherty, Using ZMET to explore barriers to the adoption of 3G mobile banking services, International Journal of Retail and Distribution Management 31 (6) (2003) 340–348. [30] H.F. Lin, An application of fuzzy AHP for evaluating course website quality, Computers in Education 54 (4) (2010) 877–888. [31] H.F. Lin, An empirical investigation of mobile banking adoption: the effect of innovation attributes and knowledge-based trust, International Journal of Information Management 31 (3) (2011) 252–260. [32] Y. Lu, L. Zhang, B. Wang, A multidimensional and hierarchical model of mobile service quality, Electronic Commerce Research and Applications 8 (5) (2009) 228–240. [33] X. Luo, H. Li, J. Zhang, J.P. Shim, Examining multi-dimensional trust and multifaceted risk in initial acceptance of emerging technologies: an empirical study of mobile banking services, Decision Support Systems 49 (2) (2010) 222–234. [34] P. Mahatanankoon, H.J. Wen, B. Lim, Consumer-based m-commerce: exploring consumer perception of mobile applications, Computer Standards and Interfaces 27 (4) (2005) 347–357. [35] N. Mallat, M. Rossi, V.K. Tuunainen, Mobile banking services, Communications of the ACM 47 (5) (2004) 42–46. [36] J.H. Marler, S.L. Fisher, W. Ke, Employee self-service technology acceptance: a comparison of pre-implementation and post-implementation relationships, Personnel Psychology 62 (2) (2009) 327–358. [37] B.F. Noble, Strategic environmental assessment quality assurance: evaluating and improving the consistency of judgments in assessment panels, Environmental Impact Assessment Review 24 (1) (2004) 3–25. [38] J.C. Nunnally, Psychometric Theory, McGraw-Hill, New York, 1978. [39] S. Percin, Use of fuzzy AHP for evaluating the benefits of information-sharing decisions in a supply chain, Journal of Enterprise Information Management 21 (3) (2008) 263–284. [40] J. Riivari, Mobile banking: a powerful new marketing and CRM tool for financial services companies all over Europe? Journal of Financial Services Marketing 10 (1) (2005) 11–20. [41] T.L. Saaty, The analytic Hierarchy Process, McGraw Hill, New York, 1980. [42] T.L. Saaty, Decision-Making for Leaders, McGraw Hill, New York, 1986. [43] T.L. Saaty, How to make a decision: the analytic hierarchy process, European Journal of Operational Research 48 (1) (1990) 9–26. [44] J. Santos, E-service quality: a model of virtual service quality dimensions, Management Service Quality 13 (3) (2003) 233–246. [45] R.M. Stair, G.W. Reynolds, Principles of Information Systems: a Managerial Approach, Course Technology, Cengage Learning, Boston, MA, 2010. [46] D. Straub, M.C. Boudreau, D. Gefen, Validation guidelines for IS positivist research, Communications of the Association for Information Systems 13 (1) (2004) 380–427. [47] A. Sulaiman, N.I. Jaafar, S. Mohezar, An overview of mobile banking adoption among the urban community, International Journal of Mobile Communications 5 (2) (2007) 157–168. [48] F.B. Tan, J.P.C. Chou, The relationships between mobile service quality, perceived technology compatibility, and users' perceived playfulness in the context of mobile information and entertainment services, International Journal of Human Computer Interaction 24 (1) (2008) 649–671. [49] E. Turban, D. King, D. Viehland, J. Lee, Electronic Commerce 2006: a Management Perspective, Pearson Education, New Jersey, 2006. [50] P.J.M. van Laarhoven, W. Pedrycz, A fuzzy extension of Saaty's priority theory, Fuzzy Sets and Systems 11 (3) (1983) 229–241. [51] P.A. Vlachos, A.P. Vrechopoulos, Determinants of behavioral intentions in the mobile internet services market, Journal of Services Marketing 22 (4) (2008) 280–291. [52] W.Y. Wu, S.H. Hsiao, H.P. Kuo, Fuzzy set theory based decision model for determining market position and developing strategy for hospital service quality, Total Quality Management 15 (4) (2004) 439–456. [53] A.S. Yang, Exploring adoption difficulties in mobile banking services, Canadian Journal of Administrative Sciences 26 (2) (2009) 136–149. [54] Y.S. Yeh, Y.M. Li, Building trust in m-commerce contributions from quality and satisfaction, Online Information Review 33 (6) (2009) 1066–1086.

204

H.-F. Lin / Computer Standards & Interfaces 35 (2013) 195–204

[55] T.K. Yu, K. Fang, Measuring the post-adoption customer perception of mobile banking services, Cyberpsychology & Behavior 12 (1) (2009) 33–35. [56] T. Zhou, Understanding users' initial trust in mobile banking: an elaboration likelihood perspective, Computers in Human Behavior 28 (4) (2012) 1518–1525. [57] D. Zhou, D. Ma, M. Efraim, Journal quality assessment: an integrated subjective and objective approach, IEEE Transactions on Engineering Management 48 (4) (2001) 479–490. [58] K.J. Zhu, Y. Jing, D.Y. Chang, A discussion on extent analysis method and applications of fuzzy AHP, European Journal of Operational Research 116 (2) (1999) 450–456.

Hsiu-Fen Lin is a professor in the Department of Shipping and Transportation Management, National Taiwan Ocean University. She received her PhD degree in Information Management from National Taiwan University of Science and Technology, Taiwan, in 2004. Her research interests include electronic commerce, knowledge management and organizational impact of information technology. Her research has appeared in Information and Management, Journal of Information Science, Behaviour & Information Technology, Internet Research, Technovation, Management Decision, and several conference proceedings.