Entrepreneurship Strategy through Social Commerce Platform: An Empirical Approach Using Contagion Theory and Information Adoption Model

: Entrepreneurship is the readiness and ability of an organization, primarily a new business, to develop, organize, and conduct its business to make a proﬁt despite uncertainties. Social commerce (s-commerce) assists consumers to buy products online. However, few studies have investigated the inﬂuence of entrepreneurship and online platform capability on consumers’ online purchase decisions. Academicians, researchers, and practitioners are also increasingly interested in understanding how the s-commerce environment inﬂuences entrepreneurship and online purchase decisions. Against this background, this study set out to examine this phenomenon. Using information adoption models and contagion theory as well as the input from the literature review, a theoretical model was developed. Such a model was tested with a factor-based PLS-SEM approach by analyzing the responses of 342 respondents. The results ﬁnd that electronic WOM (e-WOM) credibility, predicted by online e-WOM content and platform credibility, and impacted by online reputation, could signiﬁcantly inﬂuence consumers’ online purchase decisions. The study also ﬁnds that both positive and negative valance of eWOM as well as entrepreneurship signiﬁcantly inﬂuence eWOM credibility, which in turn positively inﬂuences consumers’ purchase decisions when using online platforms.


Introduction
The academic community has lately focused more attention on social commerce (scommerce), which is a subset of electronic commerce (e-commerce) that employs social media techniques to rapidly develop e-commerce [1]. S-commerce can provide a seamless shopping experience without the need for one to switch to other platforms. The internet platforms that marketers and consumers now frequently use could provide both opportunities and challenges, as consumers' processes of searching for information and making decisions have undergone paradigm shifts due to prevalence of interactive online environments.
Studies have found that consumers trust their own perceptions that are based on selfdeveloped information less when deciding on a purchase. Prospective consumers tend to seek electronic WOM for recommendations or negative information from others to mitigate their uncertainty before making a purchase [2,3]. Through internet-based s-commerce platforms, receivers of eWOM can obtain detailed information (good or bad) to help them make a purchasing decision [4]. This search for information on the internet is a challenge to marketers who are mostly focused on attracting consumers' attention.
Online sales have multiplied exponentially since the recent pandemic, and over two billion people are reported to have purchased products or services that way during 2020 [5]. It is also expected that e-retail revenue would reach USD 6.4 trillion by 2024 [6]. Thus, online channels are expected to create a popular retail environment for organizations. An influx of information helps consumers to decide which products best meet their needs [7]. Online systems can also provide consumers opportunities to experience and visualize the products with augmented reality (AR) and incur less costs when purchasing these products [8,9].
It has been observed that majority of consumers will resort to electronic word of mouth (eWOM) when making a purchase decision [10]. From analyzing the sources of eWOM, it was observed that social media sites account for a considerably high proportion (72.6%) of its sources [11]. This indicates that consumers are not only the information receivers but also information or content generators, and they share brand and product information on social media platforms [12]. Thus, communication dynamics of traditional WOM have changed as multiple interactive communication channels have emerged in an s-commerce environment that affect consumers' online purchase decision-making processes [13,14]. Several studies are available that analyzed how online consumer purchase decisions eventually help them towards making an actual purchase. However, they have not extensively analyzed how eWOM credibility impacted by both positive and negative valance eWOM could influence consumer purchase decisions through online platforms. Moreover, studies are also scant that investigated how credibility of the social commerce platforms, influenced by website reputation, and the source credibility can eventually influence consumer purchase decisions [12]. With the help of contagion theory [15] and the information adoption model (IAM) [16], the present study has attempted to investigate holistically how the credibility of eWOM and online platforms could influence consumers' purchase decisions and aimed to address the following research questions (RQs).

RQ1:
What are the antecedents that could predict eWOM credibility and platform credibility? RQ2: How can eWOM credibility and platform credibility impact consumers' online purchase decisions?

Literature Review and Theoretical Foundation
Social media platforms are considered essential places for consumers to share comments about the products and services they have used, and therefore, evaluating eWOM has emerged as an important factor in e-commerce research [17]. Studies have indicated that consumers reinforce their social networking by following other consumers on scommerce platforms, which is also an opportunity for them to share eWOM in online chat forums [18,19]. Consumers tend to seek eWOM from others' testimonials so they can be more certain about product prices before making a purchase decision [3].
Studies have also found that such sources of online information could effectively influence consumers' opinions of a brand and their purchase behavior [20,21]. Sussman and Siegal's information adoption model showed that received information and consumers' behavioral intentions are corelated [22], and this study shows that information received through eWOM impacts purchase decisions. Their model explains which factors are important for the receivers when they accept or reject information. Purchase intention is also related to information adoption, which is influenced by its usefulness [22]. This elucidates that it is important that the information is credible to affect consumers' opinions of a brand [23]. The information adoption model also supports that the behavioral intention of the information receiver is also influenced by the credibility and reputation of the platform [24].
Wallace et al. [25] and Michel et al. [26] have demonstrated that brand image as well as personal preference can determine consumers' purchase decisions. Hsu [12] examined consumers' online purchase decisions in a s-commerce environment and found that online reviews play an important role [12]. He also demonstrated that consumers' behavioral intention can be interpreted as relying on cognitive, homophily, consistency, and social capital theories. It has been observed that consumer purchase intention also depends on the IAM [27], elements of social exchange [28], and cognitive affection behavioral model [29]. This concept is in consonance with the information adoption model (IAM), which posits that individuals would adopt information from social media provided they found that the quality of the information is good and that the source is authentic and credible. IAM has dimensions like information quality, credibility, needs, usefulness, purchase intention, information adoption, and attitude towards such information.
The cognitive affective behavioral model principally possesses three components that are affect, behavior, and cognition. "Affect" indicates an individual's feeling about an object, "behavior" denotes a person's intention towards an object, and "cognition" is one's belief about an object. Thus, the affective component is concerned with feelings, behavior is associated with attitude, and cognition relates to belief and knowledge. When all these components are synchronized, individuals can make decisions.
However, there have been scant investigations on consumer behavior in using scommerce platforms to make purchase decisions and that considering the credibility of eWOM and social media platforms [12]. It has been noted that the personal characters of eWOM speakers and eWOM listeners (receivers) can impact the usage of eWOM in case of purchasing a product or service [30]. Studies have also demonstrated that listeners of eWOM rely on its credibility and its positive and negative valances [31,32]. Other studies have highlighted that online consumer purchase decisions are impacted by the credibility and reputation of the e-commerce platform that has reliable sources [33,34]. Though eWOM can be considered the basis for information transformation, those who receive the information may react differently from others [35][36][37] depending on their feelings and experiences and on how they perceive the credibility and reputation of the online platforms [38]. Other studies have considered that social media platforms are essential eWOM platforms where consumers can comment about products [8,17,39]. To collect others' opinions and to remove their doubts about products or services, potential consumers seek eWOM reviews [3,38], which play a critical role in influencing their purchase decisions [12]. However, few studies, especially focusing on s-commerce platforms, have identified what the salient antecedents are which could impact eWOM credibility, thus impacting consumers' online purchase decisions. Studies are also scant that extensively investigated, in the context of a potential consumer's purchase decision in an online environment, how platform reliability and credibility may play vital roles. Hence, there is a gap in the literature, which this study has attempted to fill. In the information system research, dual process moderation theories are being used to explain the impact of knowledge, thoughts, and information on the people who receive them [40]. Using the elaboration likelihood model, the information adoption model has been proposed for interpreting how receivers are impacted by information found in the communication-embedded environment [22,41,42]. IAM asserts that information impacts receivers' attitudes and behaviors.
Xiang et al. [43] used the information system process perspective to explore the positive and negative contributions of eWOM from online reviews. They found that when a receiver of eWOM feels that the recommendations are popular, the individual will be aligned to act accordingly. This is in consonance with the concept of contagion theory [12,15], which explains that the behavior of a receiver of eWOM reviews depends on the influence of the society to which the receiver belongs. This theory posits that people act according to the crowd's impulses. In such a context, if others find that an e-commerce website has credibility and a good reputation, the person who receives eWOM from it will be convinced and will make a positive purchase decision. Thus, in terms of contagion theory, platform credibility, depending on the source reputation, can impact consumers' online purchase decisions. Hence, by integrating the information adoption model [16] with contagion theory [15], it can be inferred that eWOM credibility, comprising positive and negative valances, and platform credibility, where the e-commerce website has a good reputation and believable sources, could impact consumers' online purchase decisions.

Methodology and Material Used
From the inputs of the literature and theories, it was possible to identify the determinants of eWOM credibility that includes positive valance eWOM and negative valance eWOM, which could impact online consumer purchase decision. Also, the authors identified that platform credibility predicted by e-commerce website reputation and electronic source credibility is perceived to influence online consumer purchase decision. All these antecedents will be explained in the subsequent sections.

Online eWOM Content (Entrepreneurial Competency)
Consumers' self-developed information could give them little reason to trust a brand. Therefore, they tend to seek eWOM to obtain others' recommendations and comments that could help them to reduce their uncertainty for making a purchase decision [2,3]. The source of information might be positive or negative, which could impact the consumers' opinions of a brand [23] and their purchase decisions [21,44,45]. Riegner [46] conceptualized eWOM as user-generated content on the internet. It provides receivers with details of the product and service information that help them when deciding on making a purchase [4,47]. The contents of eWOM information influence the receivers, and if the receivers are confident about the information, they will find it credible and trustworthy [21].
Credibility is conceptualized as the quality of being accepted or believed as real, true, or honest. It is interpreted as the power to inspire belief [19,32]. The eWOM, which provides information about a product or service, could influence receivers' purchase behavior if the information content in the eWOM is convincing, which creates credibility [44,48]. The positive or negative eWOM reviews are considered important in the context of product or service marketing [49]. It has been observed that the number of positive eWOM reviews are more than the number of negative eWOM reviews [49][50][51]. Positive eWOM reviews can enhance consumers' purchase decisions because they help to lower the risks associated with purchase decisions [4,52,53]. Negative eWOM reviews lower the brand image, which disfavors the concerned organizations and eventually increases their promotional expenditure [12,54]. Thus, eWOM, both positive and negative, is considered to influence the thoughts, feelings, views, and other emotions of the consumers receiving such information. This concept is supported by IAM [16]. Accordingly, it is hypothesized as follows.

Online Reputation (Entrepreneurship and Reputation)
A good reputation is always considered to be a forceful means towards persuasion [55,56]. The receiver of information focuses on the genuineness of the eWOM review to form an opinion about the product and make an informed purchase decision [57]. Park and Lee [24] (2009) found that consumers attach more credence to such eWOM comments posted on reputable websites. Again, if substantial information is available on the online platform, consumers will rely on its credibility to make a purchase decision [4,58,59]. It is observed that eWOM initiators who possess positive characteristics are considered more persuasive compared to those with fewer positive characteristics [60-62] (Eagly & Chaiken, 1993; Nguyen Hoang, & Tung, 2022;Yang, 2022). Incidentally, eWOM initiators are those who have previous experience with specific products or services and share their experiences on the appropriate platform, and their credibility is considered to be an important factor that impacts the correctness and accuracy of their reviews [51,63,64]. Expert recommendations are more acceptable for consumers' decision-making process as they influence them to accept the message, which is the inner concept of contagion theory [15]. Contagion theory explains that the behavior of the receivers of eWOM reviews principally depends on the influence of their society [12]. This theory also posits that the receiver of eWOM information is influenced by the impulses of the crowd. Information derives its credibility from the platform which disseminates it, and hence, purchase decisions depend on the credibility and reputation of the platform [15] (Monge & Contractor, 2003). When the content generator of eWOM is more reliable and believable, then consumers trust the eWOM reviews more [32,63,65] ( An e-commerce website with a good reputation will be highly trusted by consumers, and that impacts their decision-making process [33] ( Bart et al., 2005). In terms of the above discussions, the following hypotheses are formulated.

eWOM Credibility and Platform Credibility
eWOM review credibility is perceived to affect the attitudes of prospective consumers when deciding on a purchase [31,62]. After perusing the eWOM review comments, consumers will decide what to purchase depending on the reliability, consistency, and credibility of the eWOM reviews [32,51]. When consumers find eWOM reviews to be more consistent, reliable, and trustworthy, they exhibit a more positive attitude towards the product [4,35]. Researchers have demonstrated that consumers who access product reviews from online discussions on internet platforms could change their personal attitudes, which is in consonance with the concept of contagion theory [15,66]. When positive eWOM review comments are thought to be credible and reliable, consumers tend to also have positive attitude about their purchase decision [63,67]. Previous studies have highlighted that product reviews posted on online discussions increase potential consumers' interest in purchasing the products they read about [51,62,66]. When consumers perceive the eWOM reviews on an e-commerce website are credible, they will trust that website [68][69][70]. Again, consumers also place much importance in accepting or rejecting a product or service based on the eWOM reviews after considering the credibility of the platform on which the comments are posted [12,71]. The social media platforms lack professional gatekeeping to efficiently monitor the content posted there. The social media platform is deemed to be credible if it possesses high quality content, requests that consumers provide impartial reviews, and makes sure that the product branding is consistent [72]. All these discussions lead to formulating the following hypotheses.
The above discussions help to develop a theoretical model that is provided in Figure 1.

Research Methodology
To test the hypotheses and validate the proposed theoretical model, the partial least square (PLS) structural equation modelling technique (SEM) is helpful. As a multivariate technique, this approach helps to assess the various path relationships between the constructs [73]. PLS-SEM can also successfully analyze an exploratory study that uses a complex model. For obtaining the data to be analyzed through this technique, a survey was conducted where inputs from different respondents were gathered and quantified on a 5-

Research Methodology
To test the hypotheses and validate the proposed theoretical model, the partial least square (PLS) structural equation modelling technique (SEM) is helpful. As a multivariate technique, this approach helps to assess the various path relationships between the constructs [73]. PLS-SEM can also successfully analyze an exploratory study that uses a complex model. For obtaining the data to be analyzed through this technique, a survey was conducted where inputs from different respondents were gathered and quantified on a 5-point Likert scale which had anchors from Strongly Disagree (SD) as 1 to Strongly Agree (SA) as 5.

Questionnaire Preparation
The questionnaire (a set of questions) was prepared with input from the extant literature. The questions were written in the form of statements with some minor adjustments so they would be useful in the context of the present study. The questions were duly pretested with 11 experts for their opinions. Seven of the experts were selected from industry and four were selected from academia. The industry experts were selected because of their professional experience in the field of this study and their awareness of consumer intention to purchase items or services using online platforms. The four academic experts were selected because each had more than 10 years of research experience in the domain of this present study. Their advice and opinions helped to modify the language of the questions to make them more understandable and simpler.
After the pretest, a pilot test was also conducted with 25 respondents who were selected through the convenience sampling process. These respondents were not included in the main survey. The inputs that emerged from the pilot test helped to enhance the comprehensiveness, understandability, and readability of the questions so that the prospective respondents would not have any constraint in responding. Through all these correctional processes, eventually 36 questions were fine-tuned. Five items were identified for each of the following constructs: PVW, NVW, WER, and ESC. For constructs WOC and PLC, six items were selected. For the construct OPD, four items were identified. The measures with the sources are provided in the Appendix A.

Collection of Data
To target the usable respondents and to collect data for analysis, social media platforms were used to find a population who use different social media platforms frequently and post comments about their own purchase experiences of any product or service. Also, this study sought respondents who have frequently made purchase decisions after consulting eWOM reviews. Google Docs were used to send the survey instruments to potential respondents through different social media platforms. Here, the convenience sampling approach was used for approaching potential respondents. Furthermore, the convenience sampling technique was used because of its advantages depending on the research design [74]. This sampling technique is cost-effective and less time-consuming, and it is an easy process for subjects who can be readily available. In the absence of sampling criteria, this process allows researchers to collect data which would not have been possible otherwise. This process provides results quickly along with reliable qualitative information. All these respondents were provided with the response sheet that contained 36 questions. They were supposed to put one tick mark in one out of five options for each question. They were requested to respond within two months (June-July 2022) using online platforms. With each response sheet, a guideline was provided describing the objective of this study and to explain to them as to how to complete it. Importantly, all respondents were assured that their identities would not be disclosed. Within the scheduled time, respondents returned 356 response sheets.
To conduct the non-response bias test, recommendations from Armstrong and Overton [75] were followed. Therefore, the independent t-test and chi squared test were performed on the first and the last 100 responses. No deviation of results was noted from these two tests, confirming that non-response bias was not an issue in this study. The researchers found that 14 of the 356 responses were incomplete and did not consider them. Therefore, analysis was performed on 342 responses against 36 questions. The demographic information of 342 respondents is provided in Table 1.

Measurement Properties along with Discriminant Validity Test
The loading factors (LF) of all of the items were estimated to assess the convergent validity. To verify the validity, reliability, and consistency of each of the constructs, the average variance extracted (AVE), composite reliability (CR), and Cronbach's alpha (α) have been duly computed. The values of the different parameters were found to be within the allowable range. Table 2 provides the results. To verify the discriminant validity of all the constructs, conditions laid down by Fornell and Larker [76] need to be satisfied. For this, the square roots of all the AVEs were estimated, and all the estimated values were observed to be more than the corresponding bifactor correlation coefficients. This satisfies the Fornell and Larcker criteria [76]. Table 3 provides the results.

Common Method Bias (CMB)
The study results depend on survey data and so the chance of CMB cannot be ignored. To mitigate the risk of CMB, a procedure was initially followed. Two of the steps were the pretest and the pilot test to help make the questions in the survey simpler with an expectation that the prospective respondents would provide unbiased responses. Additionally, to verify the severity of CMB, Harman's single factor test (SFT) was conducted. The results emerged that the first factor was 24.55%, which is less than 50% that Podsakoff et al. [77] deemed to be the highest cutoff value. However, Ketokivi and Schroeder [78] noted that the SFT is not robust enough to test for CMB. Hence, a marker variable correlation test was also conducted [79]. The results of both these tests revealed that there does not exist any distinct evidence of CMB. Hence, it can be inferred that this study was not affected by CMB.

Hypotheses Testing (SEM)
Using the bootstrapping procedure and considering 5000 resamples with an omission distance of seven, the hypotheses were tested. Cross-validated redundancy was assessed, and Q 2 value was found to be 0.064, which is positive and indicates that the proposed model possesses appropriate predictive relevance [80]. To estimate the model fit, the recommendations from Henseler et al. [81] were followed and the standardized root mean square residual (SRMR) was considered as a standard index. The values emerged as 0.061 for PLS and 0.034 for PLSc, which were both found to be less than the highest permissible value of 0.08 [82]. The results highlight that the model is in order.
The β-values, along with corresponding p-values, of all the linkages were estimated. Also, the coefficients of determination (R 2 ) values of all the endogenous variables were assessed. The results are shown in Table 4 With all these inputs, the validated model is provided in Figure 2. * p < 0.05. ** p < 0.01. *** p < 0.001

Results and Discussion
The present study has formulated six hypotheses. After verification with statistical procedures, it appears that all the six hypotheses have been supported. The study results demonstrate that PVW and NVW separately impact WOC significantly and positively as the path coefficients concerned are 0.27 and 0.19 and their levels of significance are p < 0.01 (**) and p < 0.05 (*). WER and ESC separately impact PLC significantly and positively because the path coefficients are 0.21 and 0.33, respectively, with levels of significance being p < 0.001 (***) and p < 0.05 (*). We also observe that WOC as well as PLC separately impact OPD significantly and positively since their path coefficients are 0.34 and 0.41, respectively, with levels of significance being p < 0.001 (***) and p < 0.001 (***).
Therefore, for the coefficients of determination, the results of this study found that PVW and NVW simultaneously predict WOC to the extent of 52% (R 2 = 0.52). Also, WER and ESC could simultaneously explain PLC to the tune of 56% (R 2 = 0.56). The study results also demonstrate that WOC and PLC could simultaneously explain OPD as much as 71% (R 2 = 0.71), which is the explanatory power of the proposed theoretical model.
The present study has demonstrated that online eWOM comprises antecedents of either positive valance eWOM or negative valance eWOM that impacts its credibility, which eventually significantly impacts online consumer purchase decisions. This concept has been duly supported by Hsu [12] who examined the influence of eWOM reviews on beauty enterprises, in terms of contagion theory and the information adoption model. Cheung et al. [16] investigated the impacts of eWOM on online consumer communities who use online opinions to make purchase decisions. The present study has highlighted  With all these inputs, the validated model is provided in Figure 2. * p < 0.05. ** p < 0.01. *** p < 0.001

Results and Discussion
The present study has formulated six hypotheses. After verification with statistical procedures, it appears that all the six hypotheses have been supported. The study results demonstrate that PVW and NVW separately impact WOC significantly and positively as the path coefficients concerned are 0.27 and 0.19 and their levels of significance are p < 0.01 (**) and p < 0.05 (*). WER and ESC separately impact PLC significantly and positively because the path coefficients are 0.21 and 0.33, respectively, with levels of significance being p < 0.001 (***) and p < 0.05 (*). We also observe that WOC as well as PLC separately impact OPD significantly and positively since their path coefficients are 0.34 and 0.41, respectively, with levels of significance being p < 0.001 (***) and p < 0.001 (***).
Therefore, for the coefficients of determination, the results of this study found that PVW and NVW simultaneously predict WOC to the extent of 52% (R 2 = 0.52). Also, WER and ESC could simultaneously explain PLC to the tune of 56% (R 2 = 0.56). The study results also demonstrate that WOC and PLC could simultaneously explain OPD as much as 71% (R 2 = 0.71), which is the explanatory power of the proposed theoretical model.
The present study has demonstrated that online eWOM comprises antecedents of either positive valance eWOM or negative valance eWOM that impacts its credibility, which eventually significantly impacts online consumer purchase decisions. This concept has been duly supported by Hsu [12] who examined the influence of eWOM reviews on beauty enterprises, in terms of contagion theory and the information adoption model. Cheung et al. [16] investigated the impacts of eWOM on online consumer communities who use online opinions to make purchase decisions. The present study has highlighted that platform credibility, predicted by e-commerce website reputation and electronic source credibility, could also impact online consumer purchase decision. This concept has been supported by Yang [62], who investigated consumer purchase intentions in social commerce in terms of perceived cognitive efforts, and Wang et al. [83], who investigated how social network ties could influence purchases in s-commerce communities. This study has found that online consumer purchase decisions can be improved by developing eWOM as well as platform credibility. Also, eWOM credibility could be improved by enriching its content, and platform credibility could be improved by enhancing the reputation of the e-commerce website and improving the credibility of the electronic source.
In the context of s-commerce, the present study has discussed how the eWOM review comments could influence other consumers in their purchase decisions. This present study has also elaborated the need for s-commerce platforms to provide discerning product and service information to potential consumers to help them in their purchase decisions. In summary, the present study has highlighted that online platforms provide access to a wealth of precious information on products and services, rendering it easier for the consumers to find the best products and services to meet their expectations [7].

Theoretical Contributions
The present study provides several theoretical contributions to the eWOM literature. In the perspective of eWOM issues, earlier studies focused on external information sources like review ratings, source expertise, and so on [30]. The present study has effectively considered the influence of eWOM as well as social media platform credibility on consumers' purchase intention. This study has dealt with how the credibility of eWOM and e-commerce platforms could considerably influence the attitudes of prospective consumers towards deciding on a purchase. No other studies have identified the salient predictors of both eWOM credibility and e-commerce platform credibility, which could impact consumers' online purchase decisions. This study has demonstrated how positive, as well as negative, eWOM valance and e-commerce website reputation, supported by electronic source credibility, could impact the decision-making process of potential consumers mediated through eWOM and platform credibility. This is claimed as a special theoretical contribution of this study.
In this study, the conceptualization as well as operationalization of the factors of social interaction have effectively contributed towards a deeper understanding of how consumers could interact with others using different sources prior to taking a concrete decision to purchase a product or a service through the e-commerce sites. This is claimed to be a novel contribution.
The present study has been able to extend the applicability of contagion theory [15], which posits that people act according to the overall trend of the crowd which influences them. The overall concept is that people are influenced by acting according to the crowd's opinions. In consonance with this concept, the present study demonstrates that the purchase behavior of potential consumers is considerably influenced by eWOM reviews. When several eWOM reviews provide identical information through the s-commerce platforms, consumers' purchase decisions could be impacted accordingly.
This study has also extended the application of the information adoption model (IAM). Individuals may react differently to information they receive from eWOM. Their reactions can result from their judgments, feelings, understandings, and experiences. IAM posits that information has considerable influence over reviewers' behavioral processes and attitudes, which helps them to decide whether to purchase the product or not. The present study has extended the concept of IAM [22,27], which interprets the impact of information on receivers in the context of a computer-embedded communication environment [22]. IAM theorizes that information impacts receivers' attitudes and behavioral intentions. In the context of the present study, this idea has been extended to envisage that receivers will have confidence in the information if they find the source credible and the websites that contain the information reputable. Also, it is demonstrated that the platform through which such information is shared eventually influences the online consumers' purchase intentions. This study has also extended Hsu's [12] applications of contagion theory and IAM to investigate how consumers' purchase decisions are affected by eWOM in the s-commerce environment. This adds value to the extant literature.

Managerial Implications
The findings of this study provide some important managerial implications in the domain of online community management as well as marketing. They suggest that online platforms that contain consumer opinions need to have very good reputations so that they can achieve their objective in the context of eWOM reviews. It is also demonstrated that eWOM reviews can be successful if they satisfy the needs of the consumers for valuable information of the products and services they are interested in purchasing or using. This implies that platforms that provide consumer opinions should offer customizable features that could help prospective consumers develop their own discussion forums that could reduce their information search efforts.
The present study has demonstrated that e-commerce website reputation and electronic source credibility eventually help consumers to be more certain when making a purchase decision. This implies that the marketers should ensure that the websites have a good reputation by ensuring that their contents are reliable and trustworthy in appropriately describing the usability of products or services. Another measure is to ensure the reviews of the products or services include information of their different features and their proper applications. Also, it is necessary for the platform service providers to warrant that users are fully satisfied with website services and that they are easy to navigate so that potential consumers can easily search for additional information. Social media providers should also improve the design of their s-commerce sites and develop marketing strategies to increase interactional activities among the consumers. The providers should provide convenient channels to consumers for seeking as well as sharing their experience of purchase, rendering the product or service information more comprehensive. Moreover, it is suggested that the providers of s-commerce sites invest more in using social technologies to develop a simpler feedback mechanism so that potential consumers can gather detailed product information from the experience of the other consumers.

Limitations and Future Scope
Though the present study has several theoretical and practical implications, it is not free from limitations. First, the study findings depended on cross-sectional data that invite defects of causality between the relationships of the constructs and create the endogeneity flaw. To eliminate these defects, it is suggested that future researchers should conduct longitudinal studies. Second, the study results were obtained by the analysis of inputs from 342 respondents. This sample does not represent the entire society, and hence, the results lack generalizability, and future researchers could rectify this issue by considering more respondents. Third, the explanatory power of the model is 71%. It is suggested that future researchers should consider in addition other factors like brand equity [84] and interpersonal homophily [85] to examine if their inclusion could enhance the explanatory power of the model. Fourth, this study has not targeted any specific s-commerce platform. It is dependent on future researchers to examine if there are any differences between scommerce platforms when online consumers make any purchase decision. Finally, this study did not nurture the cultural aspects with respect to the location of the respondents representing different countries or regions. This could be considered a limitation of this study. It is suggested that future researchers could consider this issue to examine how the cultural aspects could affect the results.

PVW1
[3] I believe that consumers trust in self-developed information by the organizations.
(1)(2)(3)(4) (5) NVW4 [16] I believe that negative electronic word of mouth decreases the product image and consumer view on a particular product or services or an organization.

ESC1
[61] Reputation of information source of the product or service is critical.
(1)(2)(3)(4) (5) WOC6 [61] It is a common practice for many organizations to hire people who can make positive comments on the products or services of the organizations.

OPD1
[65] I believe many consumers make their purchase decision using electronic word of mouth.