Despite the recent difficulties experienced by dot-com companies, according to the Forrester report1, Business to Consumer (B-to-C) Internet commerce enjoys a steady growth rate (about 19% per year), and it is a familiar mode of shopping for many consumers [1]. Many scholars have argued that trust is a prerequisite for successful commerce because consumers are hesitant to make purchases unless they trust the seller [62], [77], [82] and [135]. Consumer trust may be even more important in electronic, “cyber” transactions than it is in traditional, “real world” transactions. This is because of some of the characteristics of Internet cyber transactions — they are blind, borderless, can occur 24 h a day and 7 days a week, and are non-instantaneous (payment may occur days or weeks before delivery is completed) — can cause consumers to be concerned that the seller won't adhere to its transactional obligations. Consequently, trust in an Internet business is focused much more on transaction processes [82], in contrast to that of traditional transactions involving brick-and-mortar stores where trust tends to be focused on face-to-face personal relationships. Quite possibly, the key to success in Internet business is the establishment of trusted transaction processes where e-sellers create an environment in which a prospective consumer can be relaxed and confident about any prospective transactions [66]. Since trust is likely to play an essential role in online transactions, it is important to identify the antecedents of a consumer's trust in the context of an Internet transaction.
In prior research, trust has been viewed through diverse disciplinary lenses and filters: economic [43], [65] and [132], social/institutional [26], [39] and [58], behavioral/psychological [47] and [70], managerial/organizational [9], [79], [112], [125] and [135], and technological [23], [27] and [96]. Trust is considered essential in exchange relations because it is a key element of social capital [98] and is related to firm performance, satisfaction, competitive advantage, and other economic outcomes such as transaction cost [9], [41] and [68] and search cost reductions [67].
Because trust has been studied through these different disciplinary lenses, previous research related to trust in the e-commerce context tends to be disjointed, case-specific, and/or loosely integrated. For example, most studies on technological trust have focused narrowly on issues of privacy, security, public key infrastructure, and other technical aspects of trust [13], [16], [72] and [94]. Some recent studies [64], [82], [115] and [117] have focused on the social and behavioral elements of trust in an e-commerce context, however these were again narrowly focused (e.g., they focused on a limited number of trust antecedents, or focused on trust in the community of sellers as a group), and therefore researchers have not yet developed a comprehensive understanding of the factors that predict consumer trust in the e-commerce context. Given the increasing prevalence of B-to-C Internet commerce, there is an urgent need to analyze an online consumer's decision-making process from a holistic standpoint which can provide an understanding of the complex and dynamic phenomena of trust in online exchanges. Accordingly, the specific research questions for the present study are as follows: What are the roles of trust and risk in a consumer's B-to-C online purchasing decision? Are they critical in B-to-C online transactions? And what antecedents can be identified that affect a consumer's trust and risk toward a B-to-C online transaction?
Since research on trust has been conducted from a variety of disciplinary perspectives, many definitions of trust have evolved. Prior research on traditional commerce focused primarily on interpersonal trust such as a customer's trust in a salesperson. Plank et al. [120] recognized that consumer trust could have multiple referents — salesperson, product, and company — and accordingly defined trust as a global belief on the part of the buyer that the salesperson, product, and company will fulfill their obligations as understood by the buyer. Similarly, in the e-commerce context [7], [11], [15], [24], [42], [62], [69], [76], [101], [103], [115], [122] and [135], researchers have tended to define describe trust as a subjective belief, a subjective probability, the willingness of an individual to be vulnerable, reliance on parties other than oneself, or a person's expectation. In our study, we will focus on the trust that a consumer has in an Internet vendor. Logically, this should include trust in the Website (e.g., www.amazon.com), the Website brand, and the firm as a whole. Accordingly, in this paper an online consumer's trust is defined as a consumer's subjective belief that the selling party or entity will fulfill its transactional obligations as the consumer understands them.
This paper provides several contributions. First, in order to uncover the role of trust, risk and their antecedents in B-to-C Internet commerce, this study develops a holistic trust-based consumer decision model to describe the decision-making process that a consumer uses when making a purchase from a given site. Second, to the best of our knowledge, most studies in the e-commerce environment have collected data concerning a consumer's successful purchasing experiences. Yet, because successful cases represent only a fraction of all consumer transaction behaviors, these past studies may have painted an incomplete picture (i.e., a biased view) of B-to-C electronic commerce transactions. Accordingly, in the present study we present a research design that enables us to examine transaction experiences that resulted in non-purchases in addition to completed purchases. In other words, we collected data from both “successful” cases and “unsuccessful” cases, and therefore can provide a more complete picture of a consumer's B-to-C decision-making process. Third, our testing of the proposed model with the Partial Least Squares (PLS) Structural Equation Modeling technique [48] provides empirical evidence that trust, perceived risk, and perceived benefit are strong determinants of a consumer's e-commerce transaction decision. Finally, the findings of this study provide several insights which should help practitioners better understand the role of trust and its antecedents in e-commerce, and ultimately add trust-building mechanisms into e-retailers' Websites.
This paper is organized as follows. The next section presents the theoretical framework for the study along with background theories that provide the foundation for the framework. The section also proposes the extended research model, referred to as a trust-based consumer decision-making model in e-commerce, with research hypotheses. The third section describes the research methodology and data collection. An analysis of results follows in the fourth section. The final section provides a discussion of the findings, and concludes with limitations and implications of this study.
Past research has recognized that electronic purchase decisions are inherently risky, and therefore trust may be an important factor in giving consumers the confidence they need to engage in such transactions [7], [38], [62], [78] and [135]. Yet many researchers have not systematically explored how trust and perceived risk may operate in combination to influence such decisions and what kinds of trust and risk antecedents play a significant role in the consumer trust-building process. In this paper, based on the net valence framework [119] we developed a trust-based consumer decision-making model in electronic commerce that recognizes that trust, perceived risk, and perceived benefit may directly influence purchase intentions and decisions, and trust may also influence purchase intentions indirectly by influencing risk perceptions. Additionally, the consumer decision-making model explores four different categories of trust antecedents that affect trust and/or perceived risk in such situations. While these represent important contributions for research, we also note that most of the predictors of trust and perceived risk identified in the model also represent factors that can be directly or indirectly controlled by vendors through Website design (e.g., information quality, security and privacy protections, third-party seals) or the conduct of business transactions (e.g., reputation). Thus, the model and results are likely to have important practical implications for merchants who wish to build their Internet business by increasing their customers' trust and decreasing their customers' risk.
5.1. Research findings
The empirical results suggest that a consumer's trust directly and indirectly affects his or her purchasing intention. A consumer's trust has a strong positive effect on the purchasing intention as well as a strong negative effect on a consumer's perceived risk. This study also provides evidence that a consumer's perceived risk reduces the consumer's intention to purchase, whereas a consumer's perceived benefit increases the consumer's purchasing intention. And finally, we found that these effects of trust, perceived risk, and perceived benefit on purchase intentions, ultimately had a “downstream” effect on consumers' actual purchase decisions. Thus, these findings validate the argument that a consumer's trust is an important factor in electronic, “cyber” transaction decision, and thereby support our extended version of the valence framework. Finally, the results indicate that trust to a large degree addresses the risk problem in e-commerce in two ways: by reducing perceived risk and by increasing purchase intentions directly (trust was the strongest predictor of the online consumer's purchase intention, followed by perceived benefit and perceived risk respectively).
All of the cognition-based and affect-based trust antecedents except third-party seals (TPS) had strong positive effects on consumer trust, suggesting that consumers consider a relatively wide variety of perceptions and observations when developing their trust in a Website and vendor. In addition, all of the cognition-based and affect-based antecedents except information quality (IQ) had negative effects on a consumer's perceived risk, indicating that consumers are also attentive to a relatively wide range of factors when assessing the risk from an online transaction. Among the cognition-based trust antecedents, it is interesting to note that consumers' perceptions of privacy protection (PPP) and security protection (PSP) both had strong influences on trust and risk. This suggests that both privacy and security are important for consumers as they shop online. That is, although logically it might seem that privacy superfluous when security is present, and security is superfluous when privacy is present, our results suggest that consumers independently value privacy and security. In sum, our findings provide strong support for our arguments that cognition-based and affect-based factors increase a consumer's trust as well as decrease a consumer's perceived risk in completing an e-commerce transaction.
One interesting finding is that the presence of third-party seals (TPS) did not influence consumer's trust but they did decrease a consumer's perceived risk. This is consistent with scholars' arguments [40], [83] and [104] that the presence of assurance seals has no significant effect on the consumer's trust and intention to purchase from a Website. However, this issue is not yet resolved in the literature, since several other scholars have argued that the presence of assurance seals has a significant impact on consumers' trust [73], [87], [97] and [142]. While our results indicate that third-party seals did not influence trust, the results still emphasize that third-party seals are an important factor in online commerce, because they impact purchase intentions and decisions by reducing a consumer's perceptions of risk. We also note another potential explanation for the lack of an effect of third-party seals on trust. It may be that consumers simply do not recognize a third-party seal as an assurance of trust. In fact, despite the fact that our sample comprised relatively active online consumers, 73.7% of our respondents reported that they were not even aware that Websites are endorsed by third parties. Considering that there is a relatively large industry centered on providing third-party seals, our results suggest that the industry should provide more consumer education about their seal services.
Interestingly, familiarity (FAM) had a strong direct influence on consumers' purchase intentions and trust as expected, however familiarity did not have a significant direct effect on consumers' perceived risk. One possible reason for this insignificant effect on perceived risk is that familiarity by its nature deals with complexity or uncertainty related to interfaces or procedures (e.g., searching and ordering products and/or services) which can reflect a vendor's competence and therefore its ability to honor its obligations (i.e., its trustworthiness), but not the presence of risk.
Finally, we found that our personality-oriented antecedent, a consumer's disposition to trust (CDT), had a significant effect on a consumer trust. This is consistent with previous studies on the relationship between trust and consumer disposition to trust [61], [102] and [117].
5.2. Theoretical and practical contributions
This study has both theoretical and practical contributions. From a theoretical perspective, the trust-based consumer decision-making model provides a holistic view of an online consumer's purchase decision-making process, incorporating the effects of the consumer's trust and perceived risk and benefits, and a range of antecedents of trust and perceived risk, and assessing the impact of these factors not only on purchase intentions about also on actual purchase behaviors. Thus, our study provides perhaps the most comprehensive understanding to date of the trust- and risk-related factors that consumers consider as they engage in online commerce. In addition, prior studies have often not adequately distinguished between trust, perceived risk, and perceived benefit, and concomitantly have not understood their relationships with each other or how they work independently or in combination to influence purchase intentions and decisions. By distinguishing among the concepts both conceptually and empirically, we believe we have provided important insights into their distinct roles in the online purchase experience.
From a practical standpoint, the results highlight several trust-enhancing factors that may guide the successful completion of electronic commerce transactions in B-to-C environments. Specifically, we identified a number of potentially important determinants of consumers' trust in a Website, and ultimately of their likelihood of purchasing from a Website, and we provided empirical evidence concerning the relative impact of each of these determinants on consumers' trust and purchase intentions. Thus, the theoretical framework and results may allow Internet retailers to better incorporate trust-building mechanisms into their Websites by focusing on the factors identified in this study. More specifically, our results suggest that since consumers' perceptions of privacy and security protection, information quality, third-party seals, and reputation are strong predictors of trust and/or risk, Internet business managers should pay particular attention to these factors in order to increase transaction volume.
5.3. Limitations and directions for future research
Future research will be needed to assess the generalizability of our findings. While our research participants reflect a fairly typical band of actual and potential Internet consumers, they may not be representative of all consumers. For example, older consumers may be even less comfortable with online purchasing due to their lack of familiarity with computers and the Internet. It is likely that for these consumers, trust and risk will loom even larger than for younger, more experienced individuals, therefore our model may be equally if not more predictive of purchase decisions for such consumers. Yet research is needed to consider whether this is so. Although our model received strong empirical support, we would also like to recognize the possibility of alternative models for understanding the relationships among the constructs examined in our study. For instance, we have proposed that trust functions as a mediator. In other words, the model assumes that the exogenous variables influence purchase intentions via their effect on trust. Research by McKnight and colleagues [101] and [102] suggests that these factors might influence purchase intentions directly, rather than indirectly via trust. Alternatively, trust could be positioned as a moderator of the relationship between perceived risk and purchase intention. In this view, trust would influence purchase intention only when the transaction is perceived as risky [98]. Given the early stage of research on the topic of trust and risk in Internet transactions, our aim in the present study was only to test the theoretical framework of the study, not advocate one particular model or framework over another. Thus, future research may fruitfully consider how these alternative models of the relationships among trust, risk, purchase intentions and decisions, and their antecedents, may complement or contradict each other, the various conditions under which the models may or may not hold, and ways in which the models might potentially be integrated.
