اندازه گیری امنیت ادراک شده در تجارت الکترونیکی B2C استفاده از وب سایت : معیار دوباره و اعتبار سنجی

Buyer concern about website security is a critical issue when it comes to maximizing the potential for electronic commerce transactions. Because perceptions of inadequacy can be a major obstacle to online shopping, many researchers have studied both the antecedents and outcomes of website security. Yet, the measures of security used in these studies are problematic. Although information systems researchers and business practitioners have conceptualized security as a multidimensional concept, published empirical studies have measured perceived security as a unidimensional construct. Exclusion of the underlying dimensions likely prevents researchers from fully assessing the impact of important dimensions of customers' perceptions of security. Here, we contribute to the methodological enhancement of this research stream by: (1) theoretically examining the nature and dimensionality of perceived security, and (2) developing and validating a multidimensional measure of this construct. The results from this study provide empirical justification for the conceptualization of perceived security as a formative second-order construct of perceived confidentiality, perceived availability, and perceived non-repudiation.

Internet technology and the variety of the resulting applications have revolutionized the way customers do business and interact with sellers of commercial products and services. In the retail industry, websites for business-to-consumer electronic commerce (B2C e-commerce) provide more accessible, easier, faster, and cheaper methods for individual consumers to conduct their retail transactions. As a result, online shopping has continued to gain popularity as a transaction medium. The growing popularity of online shopping has been accompanied by rising concerns about Internet security. In fact, consumer surveys reveal that concerns with security are the consumers' top reason for avoiding online shopping [39], [37] and [58]. Perceived security has become an important variable in B2C e-commerce consumers' decision-making model. Consequently, the future of B2C e-commerce may well depend on the selling firm's ability to manage security threats and improve consumer perceptions of Internet security [28]. This premise has resulted in perceived security becoming a major discussion and research topic among information systems (IS) professionals and academics. An extensive review of perceived security literature reveals an inconsistency between the conceptualization of security and the operationalization of the measures of perceived security in empirical studies. The literature suggests that IS practitioners and researchers generally agree that security is a multidimensional construct that is derived from several underlying dimensions (e.g., confidentiality, integrity, availability, non-repudiation). Yet, most empirical studies ignore the multidimensionality of perceived security and use measures that tend to capture only one dimension or are dominated by only one dimension of perceived security. While these studies add to an understanding of the role of perceived security in a variety of exchange environments, exclusion of the underlying dimensions prevents us from recognizing their significance, and therefore, such analyses may lack important details. The study reported here enhances the methodological rigor of IS research by: (1) theoretically examining the nature and dimensionality of perceived security, and (2) developing a reliable and valid multidimensional measure of perceived security. The more comprehensive and robust measure of perceived security allows more comprehensive testing of hypotheses related to the role of perceived security in online shopping and its impact on other endogenous variables. We begin with a background about perceived security within the context of B2C e-commerce. We then identify and describe the most significant dimensions of perceived security, which are used to develop and test perceived security as a second-order construct with first-order formative dimensions, which are themselves measured by reflective indicators [24]. We conclude by discussing implications of our findings for researchers and business practitioners, as well as limitations of this study.

This study makes two important contributions to IS research. First, we both identify and validate three important dimensions of perceived security. Compared to prior studies that use measures of perceived security that tend to capture only one dimension or are dominated by only one dimension, the inclusion of these dimensions in the measure of perceived security is more consistent with the way this construct has been conceptualized in earlier studies. Moreover, this inclusion should encourage more detailed analyses that include the impact of each dimension on other important variables in the model. For instance, previous studies have demonstrated that perceived security positively impacts customers' intention to use B2C e-commerce websites [66]. Yet, knowledge that perceived confidentiality, perceived availability, and perceived non-repudiation are valid dimensions of perceived security should reveal a more detailed understanding of how each component of perceived security impacts buyer intentions. Recognition of the major dimensions of perceived security provides researchers an opportunity to add depth to their analyses and highlight the significance of each of these dimensions for improving customers' intentions. Second, this study contributes to IS research methodology by developing and validating a robust formative second-order construct model of perceived security. The demonstrated reliability and validity of this multidimensional measure should eliminate the use of the traditionally lower-quality, unidimensional measures of perceived security that have been popular in previous studies. Moreover, we provide a process for incorporating our second-order measure into standard statistical analysis techniques. For IS practitioners, the results of this study suggest that perceived confidentiality, perceived availability, and perceived non-repudiation are important facets of perceived security, and that they play an important role in customers' decision to use a B2C e-commerce website. Collectively, they have significant impact on customers' perceived usefulness, ease of use, attitude, and intention to use B2C websites. Compared to prior studies, which use measures of perceived security that tend to capture only one dimension or are dominated by only one dimension, the inclusion of these dimensions in the measure of perceived security provides e-commerce website managers with a more comprehensive metric of perceived security. Such metric allows them to develop a richer understanding of how perceived security impacts their customers' willingness to use their websites for online purchases. Such understanding will help them pinpoint where problems with perceived security might exist and, subsequently, make strategic decisions to enhance customers' perceived security. The interpretation of our results is subject to some limitations. First, our empirical results must be considered in the context of the particular subjects included in the study. The exclusive use of Korean respondents has the advantage of excluding unwanted confounding factors resulting from cultural differences. Yet, it also has the disadvantage of reducing the generalizability of the results. Second, the use of cross-sectional data allows us to examine only a “snapshot” of the impact of various antecedents on e-commerce website actual usage. Third, the use of convenience sampling may have the downside of diminishing the generalizability of the results. Building on the advances in this paper, future studies can consider the use of longitudinal data which would reveal dynamics of this phenomenon over an extended period of time.