In this section, different theoretical perspectives on user acceptance of IS/Web site are first reviewed. They are followed by reviewing literature on dimensions of IS success model including ServQual and trust that may quantify various quality attributes of Web sites.
Many researches address the issues of IT design and evaluation to enhance usability and acceptance and try to identify the determinants of IT use and acceptance (Davis et al., 1989). User acceptance is a critical factor to determine the success of a Web design project. Both practitioners and researchers have a strong interest in understanding how people accept e-commerce Web sites so that better methods can be employed to design and evaluate Web sites and predict customers’ response. The acceptance theory seeks to extend the model of customer/user-centered design, which is advocated in usability approaches (Nielsen, 2000), from interface improvement to predictions of likely usage (Dillon & Morris, 1996). To enhance our understanding of the determining variables to explain the level of Web site acceptance, it is necessary to review different perspectives that touch on this issue: Theory of Reasoned Action, Technology Acceptance Model, and Human-Computer Interface/Usability Theory.
TRA defines the relationships among beliefs, attitudes, norms, intentions, and behavior (Fishbein & Ajzen, 1975). TRA assumes that human beings are basically rational and make systematic use of information available to them when making decisions. TRA also assumes that the behavior being studied is under total volitional control of the performer (Chang, 1998). However, the prediction of behavior from intention is problematic because a variety of factors in addition to one’s intention determine whether the behavior is performed.
TRA is a general model and may provide a common frame of reference within which to integrate various lines of inquiry. A substantial body of empirical data in support of TRA has accumulated (Bang et al., 2000; Chang, 1998; Jackson et al., 1997). TRA has broad applicability that spans behaviors in many fields including using a computer. The theory has gone through rigorous testing that shows its robustness in predicting a person’s intention and behavior under volitional control (Bagozzi, 1992; Davis et al., 1989).
However, TRA has some boundaries: (1) The behavior should be under volitional control; (2) Intent does not change prior to the performance of the behavior; (3) Intention measures should correspond to the behavioral criterion in terms of action, target, context, and time (Liker & Sindi, 1997). Free from these boundaries, TRA can be adapted to study e-commerce Web site design because most e-commerce Web site users are under volitional control and without influence from other factors such as supervisor’s impact.
TAM of Davis (1989) is the most widely cited theory in the studies on user acceptance of technology. The goal of TAM is to predict the system acceptance and diagnose the design problems before users experience the system. TAM predicts that the user acceptance of a system is determined by two factors: (1) perceived usefulness and (2) perceived ease of use. Perceived usefulness is the degree to which a person believes that use of the system will enhance his or her performance. Perceived ease of use is the degree to which a person believes that use of the system will be free from efforts.
Since Davis’s elucidation of TAM, numerous researchers have discovered that TAM yields consistently high-explained variance for why users use/accept systems. An adaptation of TRA, TAM is specifically tailored for modeling user acceptance of IS. Being parsimonious and theoretically justified, it provides a general justification for the determinants of computer acceptance and is capable of explaining user behavior across a range of end-user computing technologies. In contrast to TRA, in which the beliefs are considered idiosyncratic to a specific context, TAM’s perceived usefulness and perceived ease of use are meant to be fairly general determinants of user acceptance. Therefore, TAM is more applicable to studying the user acceptance of an e-commerce Web site than TRA because Web site design is one kind of IT project which is always carried out in a specific context. Studies on the acceptance of Web sites using TAM theory and its refinement/extension are accumulating (Eighmey & McCord, 1998; Liu & Arnett, 2000; Mathieson et al., 2001).
HCI examines all aspects of user-interface design from the high-level concerns of organizational context and system requirements to the conceptual, semantic, syntactic, and lexical levels. HCI research concentrates heavily on the concept of usability. The usability of an application refers to the effectiveness, efficiency, and satisfaction with which specific users who are performing specific tasks in specific environments can use the application. Two major characteristics of usability engineering are prototyping and redesign in seeking to maximize the usability throughout the whole product development process. Nielsen (2000) provides a large number of related usability attributes: learnability, efficiency, memorability, errors, and satisfaction.
HCI/Usability theory is the most applicable to the Web site interface design (Behbunan-Fich, 2001; Gary, 1999; Nielsen, 2000; Smith et al., 1997). The usability of a Web site is about ease to use and usefulness. For example, it answers the question — does the audience think the site is easy to use? From the perspective of usability theory, a Web site has to be developed as friendly as possible, or customers will go elsewhere. Today’s Internet users are savvier than ever, and a Web site needs to impress them if the site wants to keep them around. What impresses customers is not fancy presentation but the functionality. Such questions should be asked when developing a Web site, for example: Can users find what they need quickly and easily? Is there a site map or an index? Do pages load quickly? Is the Web site easy to navigate?
Many researchers attempt to identify IS success factors. Among them, the dimensions suggested by Delone and McLean (1992) receive the most attention. Based on Shannon and Weaver (1949) and Mason (1987), they postulate a multidimensional model of IS success. Shannon and Weaver (1949) define three communication levels: technical, semantic, and effectiveness (on the receiver). Mason (1978) re-labels it as production, product, and influence on recipient. By surveying 180 articles on IS success, Delone and McLean (1992) propose that existing measures be classified into six major dimensions: systems quality, information quality, use, user satisfaction, individual impact, and organizational impact.
The first four metrics of Delone and McLean’s IS success model can be employed to evaluate the quality and usability of an e-commerce Web site. Systems quality means availability, response time, and usability. Information quality is completeness, relevance, and ease of understanding of information. Use means the number of site visits and the number of transactions executed. User satisfaction can be assessed as repeat purchases and repeat visits. The last two metrics refers to the impact of IS on employees and firms, which may not be useful to evaluate outside customers’ perceptions on the Web site.
The above discussed dimensions include only the system aspect of IS success but overlook the human interaction aspect (Li, 1997). This deficiency can be supplemented by the factor of service quality. Pitt et al. (1995) note that most IS effectiveness measures focus on the products rather than the services and IS effectiveness will be mismeasured if IS service quality is not included. Recently, most researchers agree that a service quality measure should be a part of IS success (Kettinger & Lee, 1995; Li, 1997; Delone & McLean, 2003).
Service quality is commonly defined as the extent to which a delivered service level matches customer expectations (Parasuraman et al., 1985). Parasuraman et al. (1985) identify ten dimensions of service quality, and these dimensions are then reduced to five dimensions (i.e., reliability, responsiveness, assurance, empathy, and tangibles), which are called the ServQual instrument (Parasuraman et al., 1988). The ServQual instrument assesses the gap between what is expected and what is delivered. Pitt et al. (1995) argue that ServQual is an appropriate instrument for measuring service quality. Van Dyke et al. (1997) question the application of ServQual to IS service quality. They argue that the instrument for service perception is better than that for the difference score of the perception-minus-expectation. They propose it is preferable to use a perceptions-only method.
The ServQual instrument has been tested in IS context (Pitt et al., 1995). The instrument uses dimensions of tangibles (e.g., up-to-date hardware and software), reliability (i.e., dependable), responsiveness (i.e., prompt service to users), assurance (i.e., knowledge to do the job well), and empathy (i.e., IS has users’ best interest at heart). There are some overlaps in the conceptual domains of service and system quality. In the e-commerce context, service quality covers assurance, reliability, and empathy while system quality covers tangibles and responsiveness.
In the e-commerce context, the reliability dimension of ServQual can also be explained or replaced by the critical concept of trust. Trust is crucial in the transactional buyer-seller relationship of customers and e-vendors because of the risk and uncertainty in the online environment (Reichheld & Schefter, 2000). Trust is the expectation that e-vendors will not behave opportunistically by taking the advantage of the situation (Gefen et al., 2003). It is customers’ beliefs that e-vendors will behave in a reliable/dependable, ethical, and socially appropriate manner (Hosmer, 1995; Zucker, 1986).
Trust is defined as the subjective probability that customers believe that an e-vendor’s technology infrastructure is capable of supporting transactions (Pavlou, 2001). Pavlou (2001) posits that digital economy encourages the creation of institutional provisions to support transactions among entities that lack the traditional face-to-face context. There are many trust-building assurances provided by e-vendors, such as certification, https, guarantees, and policies. Such trust provides favorable impersonal conditions conducive to transactional success (Zucker, 1986). Gefen et al. (2003) believe that trust is critical because of the absence of proven guarantees that e-vendors will not engage in opportunistic behaviors such as unfair pricing, conveying inaccurate information, violations of privacy, and unauthorized use of credit card information.