Customer loyalty development in online shopping : An integration of e-service quality model and commitment-trust theory

Article history: Received: April 25, 2018 Received in revised format: June 25, 2018 Accepted: August 19, 2018 Available online: August 19, 2018 The aim of this study is to explore the determinants of cognitive loyalty in an online shopping environment. The study established a theoretical model by incorporating both e-service quality model and commitment-trust theory. A total of 937 responses were collected form Indian online shoppers by using the mail survey method. We assessed measurement model and structural model by using SPSS and AMOS. Study outcomes confirm that customer satisfaction, e-trust, commitment, and cognitive loyalty were strongly influenced by e-service quality and perceived value. Further, satisfaction had direct and positive influence on both e-trust and commitment but not on cognitive loyalty. E-trust had a positive impact on e-commitment and cognitive loyalty. Lastly e-commitment had a positive influence on cognitive loyalty. Based on the existing literature, there was a dearth of theoretical understanding of cognitive loyalty in an emerging economy perspective. Thus, the current study accomplished the critical theoretical gap by encompassing previous investigations. We examined the phenomenon of customer loyalty by integrating e-service quality model and commitment-trust theory in business to consumer e-commerce environment while considering e-satisfaction as a mediator, highlighting the originality and contribution of the current research to the online consumer loyalty literature.


Introduction
India's economy is one of the most attractive in the world, and it is expected to stay so in the foreseeable future.Young consumers have taken center stage and play an important role in the development of ecommerce.As Dwyer et al. (1987) report, 34 percent (440 million) of the total Indian population consumers belong to Generation-Y, and this is the most important and attractive customer segment for eretailers.In India, online retail was just negligible in 2010, and it was reached to 3 percent in 2015, and it is further estimated to stretch more than 30 percent by 2021.In India e-commerce market in 2017 was 38.5 billion US dollar and it is estimated to touch 64 billion US dollar by 2020 and 200 billion US dollar by 2026.As Garbarino and Johnson (1999) report, the total penetration of the internet retail in India's total retail market was 2.5 percent.Moreover, it is projected to touch 5 percent by 2020, and revenue was 17.8 billion US dollars in 2017 and estimated to reach 28 billion US dollars by the end of 2018.
Web shopping marketplace is developing quickly, and it has become the ultimate exciting improvement in Indian e-commerce settings.In this digital marketing war, the success of the e-commerce websites rely heavily on consumer's satisfaction, relationship commitment, and loyalty to the shopping website, a topic little researched in the developed economy but less in emerging economies like India.The term loyalty is predominantly imperative to the success of any web-based trade as it is theorized as a vital driver of post-purchase spectacles, for example, a repeated purchase from the same online shopping website, recommending others, and spreading positive word of mouth.Loyal consumers are certainly very vital to internet business survival.For the purpose, several players practice defensive or protective marketing strategies to expand their revenue and market segment by increasing consumer retention.Though conventionally, more determinations are devoted to offensive marketing tactics (Grönroos, 1982), the investigation has revealed that defensive marketing strategies could be more money-spinning through improved cross-selling, maybe at greater prices, and positive WOM message.Meeting internet customers' expectations and different demands have become a critical issue for the e-commerce players.To attract new customers and retain existing consumers and make them loyal to a specific website, e-tailers should understand customers' shopping requirements and fulfill them.Indeed, customer satisfaction is dominating marketing literature, and it plays an imperative role in managing the long-term relationship with web shopping sites.Many e-retailers are trying to understand internet customers shopping insights and formulate various marketing strategies to meet their shopping expectations.Customers are not only educated but well connected to the information and the world.Therefore, satisfaction alone may not help in developing the loyalty, but at the same time, retailers have to make sure long-term customer commitment.Etrust in the internet vendor is an essential key to build loyalty and managing the good relationship between both seller and buyer.While progresses in information technology and web development have displayed better approaches for directing business, it has also given a road to wide-scale "faceless" crimes.Many customers buy from the online platform, but they are afraid of losing their financial and transactional related information.Thus, it is imperative for e-retailers to maintain confidentiality with customer privacy and boost confidence in developing trust.Therefore, it is shown that e-satisfaction, trust, and customer commitment are significant aspects of developing customer loyalty and sustaining for the long term.
The aim of this investigation is to explore the determinants of cognitive loyalty from an emerging economy's perspective in online shopping context drawing from Indian online shoppers.Here, we employ both E-S-QUAL model and commitment-trust theory.This investigation is expected to provide imperative managerial insights about young customer shopping habits and contribute to the present limited knowledge on loyalty.This study is planned as follows.Following this introduction, the subsequent section provides the theoretical background on E-S-QUAL model and also a brief introduction to SERV-QUAL scale.Then, the formulation of a theoretical model based on broad literature trailed by the study approach and the key results.Finally, we present a discussion of the findings, study limitations, and recommendations for the future study.

E-Service Quality
The conceptualization of service quality has its hidden establishments in "expectancy disconfirmation theory".Numerous prior researchers such as Kassim and Abdullah (2008) from the marketing stream have embraced this specific theory as the foundation for service quality measurements.Principal Service Quality Model, called SERVQUAL by Lam et al. (2004), measured service quality by utilizing the "expectancy disconfirmation theory" on the five measurements, for example, Reliability, Responsiveness, Assurance, Empathy, and Tangibility.Quality has been marked in diverse ways by different investigators.As indicated by the Japanese age thinking, quality recommends "zero distortions or deformities" in the association's services.One outstanding contribution of Loiacono et al. (2002) was to offer a compact meaning of service quality.She marked service quality as an overall judgment, perceiving with the commonness of the service and illuminated it as involving valuations of the outcomes that is the thing that the purchaser is secured from service and procedure of service act.By the recommendations put forth by Singh and Sirdeshmukh (2000) situated and operationalized service quality as a distinction between customer's longings of "what they require" and their result of "what they get".Based on this conceptualization and operationalization, they foreseen a scale called SERVQUAL.This scale sets up a fundamental certainty of enthusiasm for service quality literature and has exhaustively connected in a few service settings.Analysts have built up an estimation scale to evaluate the site quality (Thatcher & George, 2004).Nonetheless, Parasuraman et al. (1985) set up scales to measure web-based shopping site service quality and utilized each of the 22 items on E-S-Qual.
Various authors have used different models proposed by many authors.For example, Woodruff (1997) used 20 items established by Lam et al. (2004) to study the loyalty in an e-commerce context.In their research, they highlighted the aggregate relationship of E-S-QUAL and outcome variables such as satisfaction, trust and loyalty were ample, but when they take proper dimension into account, then the effect was weak.When Zeithaml (1988), developed the items in the USA which is a developed country, all 22 items were adequately loaded and yielded good results.However, Parasuraman et al. (2005) separated dimensions to check the influence of E-S-QUAL dimensions on the outcome variable of likelihood to purchase.The study was conducted in China and disclosed that two out of four dimensions (privacy and system availability) of the E-S-QUAL did not influence the likelihood to purchase in an online environment.Thus, we argue that the same scale may yield different results depending on the economic condition of the country and growth of the e-commerce industry as well.The impacts of these measurements on cognitive loyalty, which we integrated into our current theoretical framework, have been less conceptualized and empirically tested.

Service Quality
Since e-commerce is in a nascent stage in India, consumer satisfaction is the most important concern.Due to the entry of new e-retailers and consumer's awareness, providing better service has become one of the most significant challenges to e-retailers.To survive in the ambitious marketplace, firms should develop their service performance.Service quality is known by both researchers and experts to contribute to consumer satisfaction and market share.The chase of online service quality has become a crucial factor for all companies that are driven by the essential to remain competitive.Moreover, service quality is the precursor of e-satisfaction and e-loyalty.The better way to improve online consumer's satisfaction is to develop the trust by meeting their shopping needs or expectations.High standard of service quality not only makes consumers stick to one particular website, but also encourage them to be loyal consumers.Even in Ribbink et al. (2004) clearly stated that the perceived service quality not only positively influences the relational variables such as trust and satisfaction but also on e-loyalty.

Perceived Value
Perceived value is nothing but the purchaser's appraisal of the proportion of perceived superiority and perceived sacrifice concerning an item/benefit, and therefore it is very crucial for improving consumer satisfaction and encouraging to repeated purchase.Perceived value is one of the critical aspects which explains consumer's commitment and repurchase intention in web shopping.The multi-directional relationship among perceived value, consumer's satisfaction, and expectations to buy or rebuy are very well reported in conventional marketing investigations.However, in Zeithaml (2000) claimed that the consumer perceived value replicates customer's assessment of the net value of the product, and therefore it will positively influence on satisfaction and repurchase behavior.Moreover, the level of consumer satisfaction will result from consumer sensitivity to the value received; perceived value has a significant influence on consumer's e-loyalty.In Mummalaneni et al. (2016) it is quoted that perceived value was a significant precursor of e-satisfaction.In Mummalaneni et al. (2016) it is described that the customers' perceived value fills in as an imperative instrument for keeping up purchasers' trust in the online website.Grabner-Kraeuter (2002) Contends that buyers' perceived value encourages the development of the purchaser's trust in the website and thus streamlines the way toward settling on buying choices.In Zeithaml (2000) it is found that neglecting to meet the shopper's expectations with respect to the value perceived from a buy may induce the customer's doubt on the website.In Hellier et al. (2003) it is claimed that consumer commitment, trust, and perceived value are the important constructs for understanding shoppers' buying choices.Thus, the bellow hypotheses were formulated: H2a.Perceived value positively influences satisfaction.H2b.Perceived positively influences trust.H2c.Perceived positively influences commitment.H2d.Perceived positively influences loyalty.

E-Satisfaction
E-satisfaction has been established as one of the variables improving the feeling of customer's trust.When online consumers are happy with an e-retailer, such satisfaction may make a belief that the eretailer is reliable or trustworthy.In a faceless business environment, satisfaction plays an important role, and it will tremendously impact on consumer's loyalty towards any particular website.In Zeithaml (2000) it is recommended that an abnormal state of consumer satisfaction can encourage the advancement of commitment inciting enthusiastic bonds with e-retailers, which may add to consumer satisfaction.Mummalaneni et al. (2016) approved the chain of connections among consumer satisfaction and loyalty with regards to online shopping.The trust improvement process in online business settings has been nearly connected with consumer's satisfaction.Once an online consumer feels that they are satisfied with online shopping experience or online transaction, then trust will develop automatically, and it will lead to loyalty.A greater level of fulfillment adds to the improvement of an abnormal state of customer trust in the e-retailer.Mummalaneni et al. (2016) stated that, at the point when online consumers are happy or satisfied with the internet shopping website, they are more prepared to interface with that website in the future, and it will result in making consumers loyal.There are a few reasons behind satisfied online consumer to transform into a faithful or loyal consumer.Hence, the researcher proposes the bellow hypotheses: H3a: Satisfaction positively influences e-trust.H3b: Satisfaction positively influences loyalty.

E-Trust
The significant or positive impact of trust on commitment and indirectly or directly on the outcome variable through the mediation of commitment is the focal start of the commitment and trust theory (Zeithaml, 2000).This significant impact of trust on customer commitment was empirically approved and validated in the past investigations.For instance, Li et al. (2006) argued that trust advance commitment since it enhances the relationship competency and addresses the focal social needs of the online customer when building associations with others.High trust always reduces perceived ambiguity between e-retailer and online consumers.Certainly, the positive relationship between trust and loyalty has been proposed by numerous investigations.Thus, we propose the bellow hypotheses: H4a: E-trust positively influences commitment.H4b: E-commitment positively influences loyalty.

Customer Commitment
Commitment is one individual's confidence that the continuing affiliation with another entity is significant and profitable, and thus it is worth making a significant effort to guarantee the persistence of this relationship always.Commitment, as the result of long-term satisfactory communications between two exchange parties, would lead one gathering to expect that no other trade accomplices would give comparative advantages to those of its present exchange party, and the accomplice would be more opposed to move to substitute exchange parties.Additionally, commitment is a bit of the endogenous method in a social exchange arrange, in light of the way that persistent exchanges increase individual substances' knowledge of the others, bringing about less susceptibility and trust in the others, consequently inciting the development of commitment between exchanges parties.Subsequently, commitment and trust are essential factors that shape the personal conduct standards of trades among people.Thus, the following hypothesis has been posited: H5a: E-commitment positively influences loyalty.

Methodology
The investigation was done in four stages.In the primary stage, an instrument was created to measure the E-S-QUAL and perceived value, to survey the consumer satisfaction, trust, and commitment, and to assess the customer loyalty towards their web shopping sites.Stage two involved the supervision of the overview instrument for the information gathering.The third stage was an appraisal of measurement modeling (CFA).Additionally, in stage four, SEM was adopted to test the hypotheses.The present investigation is primarily concerned with the understanding development of customer's loyalty toward eretailing.This examination utilized the survey technique to test the hypotheses.An instrument for the examination was established on scales utilized in the relevant literature.Items in the scale were little amended to fit the web shopping context.Face validity was assessed whereby a leading group of specialists qualified individual item.An item was retained if a high level of consensus was observed among the experts.The final instrument had 48 items which were identified with six constructs.A five-point Likert scale was utilized (1= strongly disagree and 5= strongly agree).A sum of 952 responses were received, of these, 15 were rejected in view of the incomplete survey, leaving a total of 937 substantial responses for use in statistical analysis through SPSS and AMOS software.The technique concern measures engaged and data examination follows two stages.Firstly, is to confirm the factor structure of measurement items and to establish model reliability, validity, and fit.Secondly, is to research the relationship among various constructs and test the hypotheses.

Exploratory Factor Analysis
As suggested by Zeithaml (2000) we assessed the dimensionality of the survey instrument.the 48 measurement items were scrutinized, and we set the cutoff value of 0.4 as significant factor loading as recommended by (Mummalaneni et al., 2016).We have deleted all the items which had low factor loading and low communality (<0.4).Finally, we retained only 38 scale items.Moreover, the similar method was adopted by applying principal component extraction with Varimax rotation method.This evaluation resulted in the same six factors and altogether explained 92% of the variance.The analysis outcomes indicate that all the items demonstrate communality value higher than 0.4 and factor loading greater than 0.50 (Hennig-Thurau et al., 2002).In the sub-constructs of service quality construct we conclude that only 4 items out of 8 items efficiency sub-construct load on the factor extracted from EFA, only 3 items out of 4 items system availability sub-construct load on the factor extracted, all the 7 items of the fulfillment sub-construct load on the factor were extracted, and all the 3 items of privacy sub-construct load on the factor were extracted from EFA.Moreover, we retained only four items in each remaining constructs of the proposed total items.The alpha was attained for the resulting items of the four factors of the Service Quality construct and other five individual constructs as presented in Table 1 and all the values were being well above acceptable level, and there was no item with an item-to-total correlation that fell less than 0.5 for each of these constructs.The reliability coefficient for all the constructs is summarized in Table 1.

Reliability
Reliability is a measure of strength and consistency of a measurement tool.It measures the degree to which a measurement instrument produces the same outcomes on repeated trials.It is observed that the alpha value of every latent construct is more than 0.6, which approves that all the latent variables are consistent or reliable.Correspondingly, it is found that the (CR) Composite Reliability of each latent construct is greater than 0.6, which again supports that the latent variables are internally reliable.

Convergent Validity
To ratify that all the items of a construct are indeed related to the construct, we analyze the convergent validity for each construct.The analysis outcomes show that the AVE is more than 0.50 for each construct, which shows that the items of the identical construct are correlated.Additionally, it is observed that the items of the same construct have a greater factor loading on the original construct.Thus, we can safely conclude that there is a convergent validity between the constructs.

Discriminant Validity
Discriminant validity evaluates whether concepts or measurements that are supposed to be unrelated are, in fact, unconnected.Discriminant validity was established in the measurement model examination to determine the correlations amongst the constructs and other latent variables.Diagonal elements are the square root of the average variance extracted.Off-diagonal elements are the correlations among constructs.For discriminant validity, diagonal elements should be higher than off-diagonal elements.A measurement model was constructed with all the latent variables.The CFA results show that model fit indices are above the satisfactory level.Thus we can conclude that the measurement models with all the indicators fit the data well (Grabner-Kraeuter, 2002).

Model Estimation and Hypotheses Testing
Structural equation modeling was used to assess the hypothesized model presented in Fig. 1. Results show an outstanding fit of the model.The Chi-square was non-significant (χ 2 = 2.103 d.f., p = 0.000).All other indicators also demonstrate a good fit.GFI was 0.929, AGFI was 0.98, RMR was 0.074, CFI was 0.986, and RMSEA was 0.034.All the model fit indexes surpass their corresponding common acceptable level.Thus, overall the final data shows an excellent fit for the hypothesized model.The six constructs validated using CFA was assimilated into the structural model.The model was verified adopting the SEM using the maximum likelihood estimation procedure in AMOS.SEM generated model indices, which specified an acceptable model fit as shown in Table 3. Path coefficients were analyzed to assess the several hypotheses theorized.The positive relationship of E-S-QUAL with e-satisfaction (β 0.665, p<0.001), with e-trust (β 0.544, p< 0.001), with e-commitment (β 0.598, p<0.001), and with cognitive loyalty (β 0.396, p<0.001) confirmed the findings of earlier studies specific to internet shopping.Thus, H1a, H1b, H1c, and H1d were accepted.Perceived value presented a positive relationship with satisfaction (β 0.159, p<0.001), with e-trust (β 0.160, p< 0.001), with relationship commitment (β 0.136, p<0.001), and with cognitive loyalty (β 0.178, p<0.001), hence, H2a, H2b, H2c and H2d hypotheses were accepted.Further, e-satisfaction has a significant impact on e-trust (β 0.359, p<0.001).However, it has a negative as well as the insignificant relationship with cognitive loyalty (β 0.011, ns).Hence, H3a was accepted, and H3b was rejected.E-trust showed a positive and significant relationship with both e-commitment (β 0.063, p< 0.001) and cognitive loyalty (β 0.059, p< 0.001), hence, both H4a and H4b were accepted.Finally, customer commitment has a direct and significant influence on (β 0.109, p< 0.001) cognitive loyalty.Hence, H5a was accepted.

Discussion and Conclusion
Overall, the results offer strong support for the theoretical model of relationships between E-S-QUAL, perceived value, e-satisfaction, e-trust, customer commitment and e-loyalty towards online shopping.Numerous findings are worth mentioning.To start with, E-S-QUAL was found to significantly affect satisfaction, e-trust, commitment and cognitive loyalty.As we expected, online customer assessed "service quality" as mainly relying on the internet shopping site's adequacy in searching and purchasing products, system availability, on-time delivery and maintaining customer's privacy.Individuals are likely to form favorable feelings of gratification and increase their utilization of web shopping if they admit that it will increase their shopping performance and proficiency.Second, outcomes show that perceived value has a substantial impact on customer satisfaction (β=0.159),e-trust (β=0.160),customer commitment (β=0.136) and cognitive loyalty (β=0.178).Perceived value is the most important aspect of developing loyalty in an online shopping.Customers always assess the perceived quality and perceived sacrifice of product and services.If the ratio is worth, then customers become delighted, and this will help them to commit to a specific website and form loyalty.In this regard, e-commerce players should provide the value for the money and make consumers feel happy and loyal to their shopping website.
Third, results demonstrate that satisfaction is the predominant aspect of online shopping.However, surprisingly, satisfaction does not significantly impact on cognitive loyalty, and it supports the previous study (Fornell, 1992).A possible reason for the insignificant impact of satisfaction is that customers may not be happy if only order taking and product delivery fulfilled, but they are also concerned about postsale services.If e-commerce players provide quality of service till only product reaches customer doorstep, then the job is not completed.At the same time, they should make customers delighted through handling post sales complaints promptly.Online shoppers are not certainly interested in policies, but surely in how problems are handled when they ran into difficulties.If both pre-sale and post-sale transactions are fulfilled as per the customers' expectations, then only they would think about being loyal towards a specific website.Thus, online sellers should spend their available resources to make consumer delighted and maintain a long-term relationship with them through good quality products and excellent post-sale services.
Fourth, e-trust is a noteworthy indicator of commitment and cognitive loyalty in web-based shopping.Furthermore, results demonstrate that e-trust represents 72% of the variance of commitment.It appears reasonable to propose that building trust is as huge as creating a phenomenal platform to keep up a superior relationship; besides, this is a key method for improving loyalty in the internet business.Commitment-trust theory researchers are firmly prescribed to look further into the relationship between these constructs and investigate them theoretically and empirically.
Fifth, customer commitment has a significant influence on cognitive loyalty.Thus we found a favorable relationship among commitment and cognitive loyalty.Once, customers commit and loyal to a particular website, then they will surely repurchase from the same website and recommend to others also to shop with their favorite website.In 21 st century marketing, managing the long-term relationship and retaining existing customers is the most significant challenge for the e-retailers.It would be a great idea if internet players follow "promise less and deliver more" mantra to make their customer happy and maintain a healthy relationship.Thus, online vendors should always focus on customers shopping expectations and try to fulfill within the appropriate time.In this digital marketing war, loyalty is basic to the web dealer's survival and achievement.These constructs have been broadly investigated in all around developed economies like the USA and the UK in both the online and offline setting yet determinedly less in developing economies.By incorporating both E-S-QUAL model and commitment trust theory, a richer understanding of the online customer's underlying beliefs and subsequent loyalty intentions can be expanded.Future researchers and internet business players will find our proposed model a fertile ground for advance enhancement and improvement to understand how to inspire and keep up customer's loyalty towards web based shopping.

Implications
From a theoretical point of view, current study outcomes infer that impression of quality by itself is not adequate developing loyalty.For example, quality of service is essential yet not sufficient to form loyalty. Satisfaction may act as hygiene (loyalty maintaining) factor.That is, the online shoppers could possibly be satisfied with a web store providing good services, yet she/he will certainly be disappointed with a web store providing low quality service.E-S-QUAL can contribute satisfaction to some degree, yet it is the perceived value that leads to a superior range of satisfaction.Likewise, e-trust represents an additional critical element of purchaser-seller relationship in web shopping that has been unnoticed in the literature.
The integration of commitment-trust theory additionally brings about a further descriptive model that well describes loyalty towards web shopping.Moreover, the degree of explained variance in commitment implies that is possibly among the most significant antecedent of loyalty in online sellers.Moreover, a noteworthy finding of the research is the main part of E-S-QUAL model and the commitment-trust theory in building cognitive loyalty.However, a few examinations have discovered that commitment has less of an impact than e-trust on loyalty in the web based shopping.Our outcomes recommend that the relative significance of every one of the constructs might be context specific.By and large, the research extends the loyalty literature from employee-organization relationship to customer-retailer relationships, shedding light on the commitment and trust building potential of the E-S-QUAL and perceived value.

Limitations and Directions for Future Research
Like different investigations, the current investigation also suffers from a few limitations.First, a limited sample of 937 shoppers in different product categories was considered, and this warrants carefulness in generalizing our results to any one particular category or segment.Second, the research was conducted among an exclusive urban customer base that may bias the outcomes.Third, this examination is crosssectional however, not longitudinal; the formulated causal relationships could only be conditional rather than proven.At last, this examination considered just experienced and satisfied customers, thus care must be taken before generalizing the outcomes to non-experienced customers, disappointed customers, and first time buyers.However, various elements have been considered as determinants of customer loyalty in our theoretical model, additionally study is urged to inquire about whether other possible variables like relative attitude, repurchase intention, website design, and firm reputation influence customer loyalty.Second, the examination included different e-tail brands from different product segments.Future examination can be directed by considering particular e-tail brand with particular product classifications like apparels and fashion, electronic, books and CDs, personal care, and home appliances.

Table 1
The Measures and the Psychometric Properties

Table 2
Discriminant Validity for the Measurement Model

Table 3
Results of Estimation Structural Model