Keywords

1 Introduction

Technology is narrowing the gap between physical and online shopping environment. The customers are becoming very selective in choosing the products. This is easily possible in online shopping. Consumers are keen to search online shopping, online booking, online financial transactions, etc., and their offerings.

It raises the Internet users and growth in electronic commerce (e-commerce). Enterprises are attempting to gain a competitive advantage by using e-commerce for interacting with the customers [33]. These types of businesses are commencing to realize that success of any business is not only the low price of the product, but service quality of their website is equally important. But service quality is an intangible and theoretical construct that is not easily elucidated and evaluated.

In India, the growth has been forecasted in the online retail market from 2012 to 2018 (in billion US dollars) as shown in Fig. 1.

Fig. 1
figure 1

Growth in sales in online retail from 2012 to 2018 (Source: Ref. [40])

Full size image

This shows that there is a need of research which can explore the factors affecting the service quality of online shopping sites so that sites can be compared and ranked.

2 Literature Review

Literature in domain of service quality and e-service quality has been reviewed for the present study and has been categorized as follows:

  1. (a)

    Service quality

  2. (b)

    Service quality measurement

  3. (c)

    E-service quality

  4. (d)

    E-Service quality measurement

2.1 Service Quality

Service quality has drawn a major attention of researchers; as a result, numerous studies are available in the literature. Gummesson [15] was first to suggest the concept of service quality and its strong association to the trust and perception. Oliver [25] conceptualized the service quality based on his disconfirmation model. Thus, service quality is generally considered to be a measure of how soundly the level of service delivery matched the expectation of customer.

The previous studies have also revealed that there is less managerial control over the service quality due to higher involvement of customers in the process [26]. Many researchers considered service quality as economies of services as it is playing a significant role in economic environment. Thus, Table 1 presents the select compilation of definitions on service quality.

Table 1 Definitions of service quality
Full size table

2.2 Service Quality Measurement

There are various measures available to measure the service quality; SERVQUAL is the most popular scale for measuring service quality used by various researchers among all other measures [5, 9, 32].

This area of research has been very rich in terms of basic concepts and models [13], applications [1], linkages with customer satisfaction, customer loyalty, and profitability.

The previous studies have [17, 19, 31, 34, 35, 39] well documented the critique and application in different contexts and also highlight that subject of service quality is not generic and its measurement varies with respect to specific service as shown in Table 2.

Table 2 Compilation of measures of service quality
Full size table

2.3 E-Service Quality

The studies in the field of service quality have been admired for more than three decades, but recently it has been applied to the e-commerce environment [19]. The beginning of “e-service” emerged upon the expansion of Internet applications [23]. Information technology (IT) is used by e-commerce organizations to gain a competitive advantage around the world. IT is used to elaborate the interaction with customers more friendly. It is an efficient means at minimal cost to expand a vast market share. With the increase of e-service acceptance in business environment, the significance of measuring e-service quality in the virtual world has been acknowledged.

With more use of e-services in business, many studies have been conducted to better understand its dynamics [6]. The previous studies have focused on various conceptual definitions of e-service quality [36] which have been presented in Table 3.

Table 3 Definitions of e-service quality
Full size table

2.4 E-Service Quality Measurement

SERVQUAL instrument is used to measure the traditional service quality of the company according to the five dimensions: tangibles, reliability, responsiveness, empathy, and assurance. The measurement of e-service quality emerged on the basis of SERVQUAL. For measuring the e-service quality, Gefen [12] combined the SERVQUAL five dimensions into three. But various researchers also realized that there is a need of another instrument for measuring e-service quality and SERVQUAL cannot be considered for measuring the e-service quality [27]. Since then, many researches are addressed in account of e-service quality and explored different measurements (different dimensions) for measuring e-service quality. Some of them are presented in Table 4.

Table 4 E-service quality instrument
Full size table

After an extensive literature review, an instrument called E-S-QUAL and E-RecS-Qual was developed to measure the service quality of online shopping websites by Parasuraman [27]. According to his study, service quality has provided online services an effective and efficient way for online browse, online purchase, and delivery of goods and service. In the present research, context E-S-QUAL has been adopted to measure the service quality of online shopping sites.

Taking insights and gap from the literature, the following objectives are framed:

  • To explore the factors of service quality affecting the service quality of online shopping websites

  • To propose the methodology for comparing the performance of online shopping sites

3 Research Methodology

In the present study, E-S-QUAL scale was adopted for measuring the service quality of online shoppers. The responses were collected from the respondents on 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). A total of 320 questionnaires were circulated in North India region, and 232 were received in the decided time period. Out of this, 157 questionnaires were used, and the rest were eliminated because they did not have any online shopping experience. The number of responses was considered suitable for exploratory factor analysis as suggested ratio (1:5) according to Hair [16]. The reliability of the instrument was found to be 0.698, which is to be considered acceptable according to Nunnally [24]. The value of KMO was 0.732 which shows the sample is adequate [3]. Five factors were identified on the basis of their eigenvalue and factor loading, since it is more than 1 and 0.5, respectively. These five factors were named as efficiency (F1), system availability (F2), fulfilment (F3), contact (F4), and privacy (F5). On the basis of this, expert opinions were taken to develop a methodology for making the comparison of online shopping sites by using fuzzy technique for order performance by similarity to ideal solution (FTOPSIS).

4 Fuzzy TOPSIS

TOPSIS is a multi-criteria decision-making technique which is used for ranking and comparing the alternatives among various alternatives by the numerical evaluations and calculations with respect to certain attributes/criterion. In this technique, weights will be specified for each criterion for measuring the relative importance which is felt by the decision maker. The basic principle of the fuzzy TOPSIS is that the chosen alternative should have the shortest distance from the positive ideal solution and the farthest distance from the negative-ideal solution in a geometrical (i.e., Euclidean) sense [18].

This study has considered three shopping sites as SS-I, SS-II, and SS-III. For the rating purpose of these three shopping sites, as suggested by Saaty and Vargas [29], five customers were selected as experts.

This group of five customers was asked to assign the weights to the three shopping sites. The steps for the methodology using FTOPSIS are as follows.

Step 1: Defining Fuzzy Decision Matrix

Five customers as experts are selected assessing the selection of shopping sites, since Saaty and Vargas [29] suggest that three to seven experts are suitable.

In the linguistic language, the item weightings are assessed from Table 5 by the team of decision makers and converted into TFN (shown in Table 6).

Table 5 Linguistic variable set
Full size table
Table 6 Decision makers’ judgment on item weightings
Full size table

The linguistic ratings and values (Table 7) are expressed in exact numerical values [7]. This rating of alternatives is formulated based on decision makers’ judgments as shown in Table 7 after conversion of linguistic terms into TFN.

Table 7 Decision makers’ judgments on rating for alternative (shopping sites)
Full size table

Step 2: Formulating the Complex Fuzzy Decision Matrix

The fuzzy item weightings and fuzzy decision matrix are formulated by converting the linguistic terms into TFN [Tables 6 and 7]. After TFN, convert this into complex decision matrix (Table 8) by using following formulas:

$$ \begin{array}{l} {\tilde{a}}_{\mathrm{ij}}=1/\mathrm{t}\ \left[\tilde{a}_{ij}^1+\tilde{a}_{ij}^2+\dots \dots ..+\tilde{a}_{ij}^{t}\right],i=1,2,\dots \dots ..,s;j=1,2,\dots \\[10pt] \tilde{w}=1/\mathrm{t}\ \left[{\tilde{w}}_{i}^1+{\tilde{w}}_{i}^2+\dots \dots \dots +{\tilde{w}}_{i}^{t}\right],i=1,2,\dots ..,s\end{array} $$
Table 8 Averaged frequency weightings and ratings of three shopping sites
Full size table

Step 3: Normalizing the Complex Fuzzy Decision Matrix

The fuzzy decision matrix now is normalized (Table 9) by using the following formulas:

$$ {\tilde{\mathrm{r}}}_{ij}=\left(\frac{a_{lij}}{a_{uij}^{\ast }},\frac{a_{mij}}{a_{uij}^{\ast }},\frac{a_{uij}}{a_{uij}^{\ast }}\right)\operatorname{}i\ \upepsilon\ \mathrm{B} $$
$$ {\tilde{\mathrm{r}}}_{ij}=\left(\frac{a_{li}^{-}}{a_{uij}},\frac{a_{li}^{-}}{a_{mij}},\frac{a_{li}^{-}}{a_{li j}}\right)\operatorname{}\operatorname{}i\ \upepsilon\ \mathrm{C} $$
Table 9 Normalized fuzzy decision matrix
Full size table

Step 4: Construction of Weighted Normalized Fuzzy Decision Matrix

With the normalized fuzzy numbers, now construct the weighted normalized fuzzy decision matrix (Table 10) by using the formula:

$$ {\tilde{\mathrm{v}}}_{ij}={\tilde{\mathrm{w}}}_{\mathrm{I}}{\ast \mathrm{r}}_{ij},i=1,2,3-----s,j=1,2,3,\dots \dots \dots \dots \dots, n $$
Table 10 Weighted normalized fuzzy decision matrix
Full size table

Step 5: Calculate the FPIS and FNIS

$$ \begin{array}{l}{\mathrm{A}}^{\ast }=\left[\left(1,1,1\right),\left(1,1,1\right),\left(1,1,1\right),\left(1,1,1\right),\left(1,1,1\right),\left(1,1,1\right),\left(1,1,1\right)\right]\\[7pt] {}{\mathrm{A}}^{-}=\left[\left(0,0,0\right),\left(0,0,0\right),\left(0,0,0\right),\left(0,0,0\right),\left(0,0,0\right),\left(0,0,0\right),\left(0,0,0\right)\right]\end{array} $$

Step 6: Calculate the Distance (Tables 11 and 12)

The distance for each alternative form FNIS and FPIS can be calculated according to the following formulas:

$$ {d_j}^{\ast }={\sum}_{i=1}^sd\left({\tilde{\mathrm{v}}}_{ij}{\tilde{\mathrm{v}}}_i^{\ast}\right)\operatorname{}j=1,2,\dots \dots s $$
$$ {d_j}^{-}={\sum}_{i=1}^sd\left({\tilde{\mathrm{v}}}_{ij}{\tilde{\mathrm{v}}}_i^{-}\right)\operatorname{}j=1,2,\dots \dots s $$
Table 11 Distance measurement of alternatives
Full size table
Table 12 Closeness coefficient (CC) of shopping sites
Full size table

Step 7: Calculate the Closeness Coefficient of Each Alternative Using the Following Formula

$$ {\mathrm{CC}}_i=\frac{d_j^{-}}{d_j^{\ast }+{d}_j^{-}}=j=1,2\dots .n $$

According to the closeness coefficient, three online shopping sites can be ranked as SS-II > SS-III > SS-I, from highest to the lowest. The results indicate that SS-II is outstanding one. SS-I performs worse than SS-II and SS-III. So SS-I has some more gaps to improve their service quality to their customers while purchasing online through his sites.

5 Discussion and Conclusion

In today’s scenario, the uses of online shopping sites are increasing day by day; thus, it requires more improvement in their services in an effective and efficient way. Improved service quality can reduce so many problems. In this paper, an integrated approach of EFA- FTOPSIS and E-S-QUAL scale is presented to evaluate the service quality, by the online shopping sites, which they are providing to their online customers. The present methodology can be applied for ranking more online shopping websites. Online companies can also use this methodology for comparing their performance with their competitors and hence make enhancement in providing the services to their online customers.

6 Limitations and Future Research

The practical difficulties have confined the study only for measuring the service quality of online shopping sites; future research may focus on the measuring quality of service in other service sectors. The dimensions and items of service quality have been only taken from E-S-QUAL. These items and factors/criterions can be varying. Another limitation for this research is number of respondents. Future research can be conducted by more number of respondents to get more generalized result.

As this case has been solved by using Fuzzy TOPSIS approach, so it is recommended that future research can be conducted in the same area by considering more items and factors with same methodology or by some other approach like analytical hierarchy process (AHP), analytical network process (ANP), etc.