Trust in e-commerce sites not only increases purchase intention directly, but it also reinforces the positive relationship between website appeal and purchase intention while attenuating the positive relationship between product appeal and purchase intention. Service content quality, search delivery quality, and enjoyment are confirmed as positive antecedents of website appeal whereas diagnosticity and justifiability are established as positive antecedents of product appeal.
This study not only delineates product and website appeal as complementary drivers of consumer purchase on e-commerce sites, but it also derives five signals that aid in bolstering both product and website appeal. Trust is revealed to exert a moderating influence on the impact of product and website appeal on purchase intention.
Practitioners should prioritize their resource allocation to enhance qualities most pertinent to product and website appeal. E-commerce sites should offer product-oriented functionalities to facilitate product diagnosticity and reassure consumers of their purchase decisions.
To do so, they can either increase the number of product variants or offer a wider array of product configuration possibilities. Product configuration makes product management more complicated, introducing optional product features and different options as well as rules and dependencies between these. Often the aim is to offer the customers the possibility to tailor their own product solution, yet with a limited number of options and from a limited, pre-defined product offering.
This is referred to as mass customization. In addition to a specific product that the customer is considering, companies also want to suggest and introduce the customers to other related products. These can be spare parts, optional or required add-ons and extensions, recommended accessories frequently purchased together with the main product, or alternative products. Alternative products can be slightly different product variants cross-sell or somewhat better products up-sell.
Sometimes there is a need to bundle products to create structured entities, special offer packages or campaigns. Ever more often, companies want to build tailored offerings to meet the needs of specific customers or market segments. Because e-commerce mainly relates to the transport of packages, bundling of goods flows — on which many of the benefits of PI are based — is particularly relevant. A core aspect of the PI is the multi-layered network structure, with different types of hubs where goods can be stored and transhipped.
Micro hubs, small-scale logistics operations at fixed locations with a small geographical coverage city center, residential area or street play an important role in this. The PIONEER project focuses on the development of PI-based concepts around micro-hubs for e-commerce, whereby a balance is sought between service orientation, efficiency, sustainability and quality of life.
The smallest type of micro hub is the street hub. This is a distribution point for parcels that can be located in a residential home.
ViaTim is a start-up that is setting up such street hubs. Parcels from web stores such as wehkamp. The street hub forms an extra link in the current logistics structure. Other variables explained in this study were well-defined in literature. Furthermore, semi-structured interviews were conducted from experts, having different backgrounds in this field of study.
Literature and interviews helped us to define the primary constructs platform interactivity, relational, and transactional contracts and were the primary source to authenticate the content validity of the questionnaire. This questionnaire was specially designed for buyers having online experience in China. Having support from literature, The English version of the questionnaire was first developed and then translated into Chinese.
To ensure the accuracy of the translation, the forward and backward translation method was applied for all constructs [62]. For the ultimate validation of the questionnaire, a pilot study was additionally conducted with the help of 25 respondents having immense experiences in online shopping. After successful pilot testing, some minor changes in structure and questions were made to reduce the cultural and language differences and to improve the reliability and validity of the questionnaire.
The study gathered data by providing an online survey link to intensive consumers of social commerce sites. The sample population was comprised of students and instructors having the online experience, aged 18 and above, from different colleges and universities in China.
This convenient sample approach within the context of online shopping was in accordance with prior researches in information systems [31]. Since the data was collected by providing an online link to the sample, so this study also checked the nonresponse biases. Nonresponse bias arises when there is a substantial difference in opinions between the actual and potential respondents [3]. Nonresponse bias refutes the outcomes of research because the survey sample cannot accurately represent the population.
Forty-six earliest and 46 last responses were compared using Chi-square tests and T-test on context variables to identify the nonresponse bias. Weitao an extension of TaoBao having both social and e-commerce features were used for this study. A total of respondents were surveyed online, were male Most of the respondents were aged between 18 and 30 Detailed demographics of the respondents are described in Table 1. Table 1: Demographics of respondents. In this research, data were analyzed by using SmartPLS version 3.
This study used the partial least square-structural equational modeling PLS-SEM method, a popular tool to examine the new research trends to build a model rather than confirmation [66].
This method helps address the foregoing limitations within the constructs, by measuring both reflective and formative constructs at the same time; therefore, PLS-SEM is the best fit over covariance based-structural equational modeling CB-SEM [15]. Furthermore, the choice of PLS-SEM was made on the base of its ability to estimate causal relationships among all latent constructs simultaneously while dealing with measurement errors in the structural model [17].
Keeping in view some limitations of PLS-SEM, like causal symmetric relationships and net effects, this study also tried to overcome these limitations by employing a new technique fsQCA, which is considered a valued algorithm for solving uncertainty issues [48].
This study evaluated the measurement model through reliability, average variance extracted AVE , discriminant, and convergent validity using commonly accepted guidelines. Reliability was assessed first, and the recommended threshold of 0. These results of reliability and AVE are shown in Table 2. Data was gathered through a single source, and it was cross-sectional, so Harman's single-factor test was used to check the common method variance CMV estimations [47]. Spurious covariance shared among variables has been tested since a common method is used in data collection.
The single factor did not account for the majority of the variance, which indicates that the common method bias did not affect the data. This study also employed the approach by Liang [32] to examine common method bias CMB. The average of Ra2 was considerably higher than their method variances Rb2 with a ratio of Hence, CMB remained not an issue. Table 2: Reliability, AVE, and correlations. Bold items are the square root of the AVEs. Validating higher-order formative factors requires a bit more finesse.
Following Marakas et al. First, this study assessed the significance of the indicators. In the case of these constructs, we included the effects from first to second-order factors.
The calculated results provided reliable provision to us for the four dimensions as traits of PI as a second-order construct in Table 3. Table 3: Upward dimension effects for platform interactivity. Second, we also have executed further tests to check the severity of multicollinearity. All constructs have variance inflation factors values less than 3, which are within the threshold of 3. The major paths from characteristics to behaviors to outcomes are all significant at the The model explains that PCs, antecedents, and consequences impact Figure 2: PLS results of research model.
H2 examined the effects of platform interactivity on relational contracts RC. H3 examined the outcomes of TCs on purchase intention.
H4 examined the effects of RC on purchase intention. Table 4 and Figure 2 show all the hypothesis results. Table 4: Hypothesis results. Furthermore, effect size f2 was examined to inspect the nominal effect of the model [10]. To test for the mediating role through H6 and H7, we employed the latest conventions [21], [34], [75], focused on bootstrapping.
For mediating effect, the indirect effect must also be significant [15]. After analyzing the model, we found that PI significantly positive had a high and significant effect on RC, which in line has a robust relationship with purchase intentions. The indirect effects of PI through the mediator RC were both significant. The same was the case recorded with ratings and reviews. The indirect effect of rating and review through the mediator TC was significant and positive.
It concludes that partial mediation exists see Table 5. Based on a thorough examination, it is inferred that both models i. Also, the outcomes show that the proposed model of this study has noteworthy predictive relevance and explanatory power. Table 5: Mediation effect of the model. The two main advantages of using FsQCA over traditional techniques are equifinality i.
Besides, fsQCA leads to a more detailed understanding of the conditions under which the outcome occur than do the regression analyses. Moreover, from a mathematical point of view, fsQCA is equally conclusive to analyses small or large sample sizes, making it a suitable tool for a wide range of research [71]. For analysis, 30 random cases are selected from the dataset.
In the fuzzy dataset, the values lie between zero and one, so calibration of data is essential. Calibration helps to generate a fuzzy-set score that relates to the degree of membership in a set. Table 6 shows the cut-off points for each of the conditions and outcomes.
Analysis of necessary conditions is the first step in fsQCA. Usually, if the consistency score is higher than the threshold of 0. Table 6 explains this analysis and elaborates on the presence or absence of outcome variables. Other conditions are not necessary for the outcome variables because they are less than the threshold of 0. Following Ragin [48], a truth table analysis is also conducted for sufficiency test to attain the possible configuration to achieve the outcome.
In this study, a consistency value of 0. The results of sufficiency analysis given in Table 7 demonstrates the possible causal configuration leading to the outcome variables. The same is found in the case of RR and the platform interactivity to generate a transactional and relational contract, respectively. These findings provide further evidence in favor of Hypothesis 1 and Hypothesis 2.
Specifically, rating and reviews, platform interactivity, transactional contract, and relational contract appeared to be the best possible configuration influencing purchase intention. The results indicate that the model is informative, as all the causal conditions are necessary to explain the outcome. Moreover, these results are very much in line with those obtained from SEM estimations, providing further evidence to believe the idea that psychological contract is essential to determine the purchase intention.
Discussion including main findings, theoretical and managerial implications, limitation, and future research as presented as follow. First, the study validates that social commerce platform interactivity is composed of four multidimensional constructs: content usefulness, engaging, human-human, and real-time communication. Though the previous research in the context of e-commerce has addressed platform interactivity, it remained limited only to one or two dimensions rather than considering all of them at the same time [12], [27].
Barreda et al. This study provides further evidence that platform interactivity should be used as a multidimensional construct to fully understand its nuances, making it a powerful construct for practitioners. Second, the results showed a significant impact of platform interactivity on relational contracts. This result is consistent with existing literature that platform interaction significantly impacts the online relations [70] and relational contract in this study.
Furthermore, platform interactivity was viewed as an essential construct to establish long-term relationships with its consumers [42].
Thus, an online platform with good interactivity can establish a favorable relational contract with consumers, thereby acquiring the ultimate goal of purchase intention. Third, the model showed that rating and reviews exhibited a positive effect on TCs. Rating and reviews are the main sources of getting information from actual consumers and also serve as a guideline about the fulfillment of the contractual obligations of the platform for potential consumers.
Fourth, the model and results showed partial mediating effects of relational contracts that impact the relationship between platform interactivity and purchase intention. TCs had also shown partial mediation in rating and reviews and purchase intention.
Fifth, we found that TCs had a significant positive effect on purchase intention in the social commerce context. The same was the case with the relational contract. Higher the perceptions of relational obligations fulfillment higher will be the purchase intentions. Fuzzy set qualitative comparative analysis fsQCA and Regression analysis are developed on two distinctive logics. Moreover, fsQCA may deal with greater leverage than regression techniques in identifying the various factors regarding configurations of conditions and their connections with purchase intentions.
Furthermore, fsQCA assists in categorizing parsimonious combination of factors that explains high purchase intentions in this study. So, fsQCA elaborates different conditions or a set of conditions that are necessary. This study contributes to the literature in the following ways. These specific advantages assist the platform with increased consumer purchase intentions, which are inconceivable without psychological contracts.
It also expands the psychological contracts as a crucial extra variable of customer platform relationships in social commerce sites, whereas psychological contracts have already been applied in e-commerce sites [13]. Second, this investigation has improved the comprehension of the psychological contract in social commerce by incorporating platform interactivity and rating and reviews. By fusing platform interactivity in a consumer-platform relationship, this study helps better to comprehend consumers' and platforms' mutual expectations from each other that help enhances online purchase intention.
Third, it delivers empirical support for the proposed model of psychological contracts and platform interactivity with its multidimensional constructs in the social commerce context. Fourth, the psychological contract is reached out as an intervening variable in this examination. It clarifies consumers' exchange conduct in social commerce sites in a more exact manner. Concerning social exchange theory [6], reciprocal obligations are demonstrated as a vital factor for initiating a first psychological contract, which affects the progress of the consumer-platform relationship.
Our study suggests social commerce platforms to prioritize the fulfillment of the psychological contracts for its loyal consumers first, who later become the power source of attracting new consumers with their reviews and ratings. New consumers may rely on the ratings and recommendations of old consumers for information before making any purchase decision. Furthermore, platform interactivity fulfills an influential role in enhancing purchase intentions.
Instantaneous communication and human-human interactions are precisely the most relevant indicators for building a robust consumer-platform relationship. As a result, the platform must initially focus on its interactive features to capture consumers' attention through its useful content and engaging tactics. In this regard, a platform should provide a transparent and user-friendly interface helping consumers to navigate and interact effectively.
These distinct features may ease potential consumers influencing their decision-making process. Despite the contributions of this study, some research limitations need acknowledgment.
0コメント