Trust and customer satisfaction in online shopping a study in vietnam

TRUST AND CUSTO M ER SA TISFA CTIO N IN ONLINE SHOPPING: A STUDY IN V IETN A M Nguyen Tlti Tuyet M - Nil am Phong Tuan Introduction The appearance o f the Internet has being paving the way for the rapid growth of electronic commerce (e-commerce) The economy and transaction methods turn to the new paee since the high-technology systems are exploited into applications Finding partners and customers is not limited by the borders o f countries and therefore the choices o f products/services increase due to more suppliers in all over thí world that are available on the Internet Beside more opportunities, the competition among electric vendors (e-vendors) has been also stronger and stronger, especially for the emerging markets in which there are many international giants Hence marketers have tried to keep customer intention by raisins customer satisfaction mainly through improving trust One approach online companies can is to ensure distributive fairness and procedural fairness Distributive and procedural fairness will trigger the feelinss of equity o f outputs (what is received) departed from inputs (what is invested) (Adams, 1963, p 347, 1965) and o f outcome-determining procedures (Folger & Greenberg, 1985) From then, trust and customer satisfaction will be maintained (Chiu Lin, Sun, & Hsu, 2009) Other aspects are customer interface quality, perceived security and perceived usefulness In the offline commerce, face-to-face interaction may directly satisfy buyers through supporting services In e-commerce, salespeople interact via interface o f the websites The challenges facingCP the online sellers are to alleviate the uncertainty and incomplete or distorted information (Ba & Pavlou, 2002) as well as ensure the security for sensitive contents and transactions Moreover, in the errerging markets, the belief o f customer on the virtual transactions is not strong Therefore, the mission o f designers is creating the attractive interface, updating latest information, and security systems, thus enhancing the perception of usefulness o f customers However,7 few studies investigate the above mentioned o cognition related to determinants o f trust and satisfaction in online context The TS Đại học Hiroshima - Nhật Bản 462 TRUST AND CUSTOMER SATISFACTION above reasons motivate our work to profoundly understand the impacting factors on trust and customer satisfaction Research m odel and hypotheses developm ent The proposed model is shown in Figure Figure Research model Distributive fairness Distributive fairness, also called as perceived fairness o f outcomes, was startỉd by Adams (1963) Adams emphasized that there are correlations between inputs aad expected outcomes The expectation departs from the contributions to the exchange, for which the fair return will be hopefully gained There are many previous studies that mention the relationship betwein distributive fairness and trust Pilai, Williams, & Tan (2001) had their strong argument on high levels o f trust ensuing fair outcomes distributions Particularly in the case ofecommerce, Chiu et al (2009) added the ideas that when customers get the products equal with their expectation, the level o f their trust in the vendor will raise On the other hand, distributive fairness is also found to be correlated with customer satisfaction Distributive fairness is traditionally explored as a predictor for customer satisfaction (Huppertz, Arenson, & Evans, 1978) In e-commerce context, Chiu, et al (2009) also tested successfully the im pacts o f distributive fairness on customer satisfaction Thus, based on the above discussion, we propose the following hypotheses: Hypothesis I (H I): Distributive fairness positively influences trust in online shopping Hypothesis (H2): Distributive fairness positively influences custoixer satisfaction in online shopping Procedural fairness Another stream o f fairness is procedural fairness which refers to the equity of the process o f how outcomes are determined (Folger & Greenberg, 1985) The relationship between procedural fairness and trust is found in many studiỉs According to Pearce, Biglev, & Branyiczki (1998), trust as well as organizational commitment results from procedural fairness in coworkers In online shopping context particularly, Chiu, et al (2009) argued that the perceived fairness o f polices and procedures o f shopping in the virtual markets has the influences on trust 463 VIỆT NAM HỌC - KỶ YẾU HỘI THẢO QUỐC TẾ LẦN T H Ử T On the other side, the coưelation between procedural fairness and customer satisfaction has been estimated Prior scholars emphasized the importance of procedural process in which the receivers not feel satisfied even though they set favorable returns In contrast, they are happy with fair procedures even if the outcomes are not proportional (Lind & Tyler, 1988) M any researches also find the positive influence o f procedure on customer satisfaction in service encounters (Bolton, 1998), in service quality (Smith, Bolton, & Wagner, 1999) and also in online shopping (Chiu, et al., 2009) Therefore: Hypothesis (H3): Procedural fairness positively influences trust in online shopping Hypothesis (H4): Procedural satisfaction in online shopping fairness positively influences customer Customer interface quality Customer interface quality is a concept involving many aspects and is measured in different ways Based on the prior researches (Chang & Chen, 2009; Thakur & Summey, 2007), our study is composed o f information and character of website The most important determinant o f e-trust is the information presentation on the website (Thakur & Summey, 2007) According to Hoffman, Novak, & Peralta (1999), customers may not trust website providers due to their suspicious entity data The online storefront design actually improves store traffic and sales, and then customer satisfaction (Lohse & Spiller, 1999) Montoya-W eis & Voss (2003) recognized that information content, navigation structure, and graphic style are three website design factors impacting customers’ use o f an online channel and overall satisfaction Therefore: Hypothesis (H5): Customer interface quality positively influences trust in online shopping Hypothesis (H6): Customer interface quality positively influences customer satisfaction in online shopping Perceived security Perceived security refers to the belief o f customers about the safely transmitting sensitive information (Chang & Chen, 2009) Hoffman, et al (1999) proved that 69 percent o f web users did not give any data to the websites because they not know what they will with the sensitive information The trustful 464 TRUST AND CUSTOMER SATISFACTION relationship between customer and e-vendor alliance o f information technology, financial 2000) In line with the discussion above, Jin (2000) and Park & Kim (2003) proved that contributor to trust and satisfaction Therefore: is build only by ensurina a major control and audit functions (Keen, & Park (2006), Szvmanski & Hise perceived security is a significant Hypothesis (H7): Perceived security positively influences trust in online shopping Hypothesis (H8): Perceived security positively influences customer satisfaction in online shopping Perceived usefulness Perceived usefulness is the belief o f customers about enhancing online transaction performances (Chiu, et al., 2009; Davis, 1989) Since customers have perceived usefulness, they will trust the e-vendor (Pavlou & Fvgenson, 2006) Perceived usefulness is essential in shaping consumer attitudes and satisfaction with e-commerce channel (Devaraj, Fan, & Kohli, 2002) The usage of Internet-based learning systems relied on extended version o f the technology acceptance model (TAM) is perceived to be useful in helping increase learners’ satisfaction (Bhattacherjee & Premkumar, 2004; Saade & Bahli, 2004) Therefore: Hypothesis (H9): Perceived usefulness positively influences trust in online shopping Hypothesis 10 (H10): Perceived usefulness positively influences customer satisfaction in online shopping Trust Based on social exchange theory (Blau, 1964), some scholars theorized that trust will create the strong impacts on customer satisfaction (Chiou, 2003; Singh & Sirdeshmukh, 2000) M organ & Hunt (1994) indicated the key role o f trust to influence customer satisfaction Singh & Sirdeshmukh (2000) specified trust mechanisms in cooperating and competing with agency mechanisms to know the effect on satisfaction in individual encounters They proved that trust will have direct effect on post-purchase satisfaction Chiou (2003) and Lin & Wang (2006) argued that accumulated trust will impact on overall satisfaction In terms of ecommerce, Chiu, et al (2009) proved that trust is the strongest variable that have impacts on customer satisfaction in online shopping Therefore: Hypothesis ỉ l (H I I): Trust positively influences customer satisfaction in online shopping 465 MỆT NAM HỌC - KỶ U HỘI THẢO Q l í ó c TÉ LÀN TH Ử TƯ Research m ethodology Data collection The data was collected in October 2011 via an online survey because of advantages o f cost and speed This online data collection method was also in consistent with the research subjects o f the study, online buyers We distributed the Ink through the survey website www.nothan.vn The duration o f survey was two nonths The participants were volunteers who were interested in such a research tipic and had the shopping experiences before A total o f 1,025 responses were received 267 out o f 1,025 responses were hvalid incomplete or save the same rating for all items; the remaining 758 cuestionnaires were used for the analysis cuestionnaires was summarized in the Table The demographic Table I: Dem ographic profile (N = 758) Frequency % Male 222 29.3 Female 536 70.7 25 90 11.9 Junior high school 0.1 High school 16 2.1 Vocational school 17 22 Technical college 39 5.1 University 676 89.3 Master or higher 1.2 Full-time student 380 50.1 Non-full-time student* 197 26.0 Employment 171 22.5 Characteristics Gender Age Education background Job Student 466 profile of TRUST AND CUSTOMER SATISFACTION Unemployed 0.8 Housewife 0.3 Retired 0.3 year 64 8.4 2-5 years 474 62.5 5-10 years 217 28.6 Years o f experience with the Internet 0.5 Ỉ years+ Number o f visits for last six months < time 81 10.7 time 417 55.0 times 148 19.5 3-5 times 85 11.2 6-10times 10 times 19 2.5 1.1 *Despite of working with permanent full-time jobs, they are enrolling some course CO have higher degrees Source; author Measurement The questionnaire was designed to measure research constructs using multiple-items scales adapted from previous studies that reported high statistical reliability and validity Each item was evaluated on a five-point Likert scale ranging from (1) Strongly disagree to (5) Strongly agree Data analysis 4.1 Analysis o f the measurement model Confirmatory factor analysis (CFA) was developed for measurement model in order to establish the unidimensionality, reliability, convergent validity and discriminant validity The good-of-fit indices satisfied the suggested value (x2/di = 2.759; GFI = 94; CFI = 0.97; TLI = 0.967; NFI = 0.956; AGFI = 0.919; RMSEA = 0.048), therefore there was a reasonable overall fit between the model and observed data The reliability assessment was based on comparable fit index (CR) All CR indexes o f constructs were over the respective recommended cut-off levels o f 0.7 In term o f convergent validity, all standardized regression weights are higher 467 VIỆT NAM HỌC - KỶ YÉƯ HỘI THẢO QUỐC TÉ LÀN TH Ứ TU than 0.60 and the critical ratios are significant at p = 0.001 In addition, two criteria, C'R and average variance extracted (AVE), were above the suggested levels, 0.7 and 0.5 respectively Finally, discriminant validity was examined using the guideline in the research o f Fornell & Larcker (1981) The correlations among constructs were listed with the AVE on the diagonal All diagonal elements were larger than inter­ construct correlations; hence discriminant validity was proved 4.2 Analysis o f the SEM Table shows the result of the SEM Referred to the corresponding recommended values all fit indices achieved a good model fit (x2 = 479.036 (p = 0.000); d f = 168; xV df= 2.851; GFI = 0.942; CFI = 0.973; TLI = 0.967; NFI - 0.96; AGFI = 0.92; RMSEA = 0.049) The explanatory power of the research model was shown in Figure in which the model accounts for 71 and 72% of variance (R score) Ten out o f eleven paths were significant Among them, nine exhibited a pvalue o f 0.001 H I, H2 were supported by the significant coefficient paths from distributive fairness to trust and customer satisfaction o f 0.232 and 0.145 Procedural fairness was associated with trust and with insignificant coefficient path with customer satisfaction, therefore H3 was supported but H4 was not supported H5, H6, H7, H8 proposed that customer interface quality and perceived security would positively impact on trust and customer satisfaction, and the results were strongly supported (7 = 0.285; p 32 = 0.165 ; 741=0.161; [3,42= 0.099) H9 and H10 posited that perceived usefulness would positively affect on trust and customer satisfaction, the result were significant, and therefore H9 and 1110 were supported H1Ỉ was supported because trust had positive influence on customer satisfaction 0*2=0.32) Table 3: T he result of the SEM Hypothesized relationship P a m e te r Estimate Critical ratio Conclusion (t-value) HI - Distributive fairness —>Trust -— Yu 0.232 5.78* Supported Distributive fairness p 12 0.145 3.85* Supported H2 —►Customer satisfaction H3 Procedural fairness —>Trust 721 0.265 6.39* Supported H4 Procedural fairness P22 0.065 1.68 Not supported —♦Customer satisfaction H5 Customer interface quality —►Trust 468 73! 0.285 7.88* Supported TRUST AND CUSTOMER SATISFACTO'J H6 Customer interface quality —♦Customer satisfaction P32 0.165 4.68* Suppored H7 Perceived security —>Trust 741 0.161 6.30* Supị:or:ed H8 Perceived security —>Customer satisafction p42 0.099 4.1 1* Supfored H9 Perceived usefulness —->Trust 751 0.091 2.94** Supported H 10 Perceived usefulness —►Customer satisfaction P52 0.179 6.34* Supported HI Trust -^Customer satisfaction ^-62 0.320 6.20* Supported Overall goodness-of-fit indices X2 = 479.036 (p = 0.000); df = 168; x2/df= 2.851 GFI = 0.942; CFI = 0.973; TLI = 0.967; NFI = 0.96; AGFI = 0.92; RMSEA = 0.049 Note: X2- chi-square; df, degrees of freedom; GFI, goodness-of-fit index; CFI, comparable fit index; TLI, tucker lewis index; NFI, normed fit index; AGFI, adjusted goodness of fit index; RMSEA, root mean square error of approximation; *p< (.031, **p