چرا ما به فن آوری های جدید اعتماد می کنیم؟ بررسی شکل گیری اعتماد اولیه با سیستم های اطلاعات سازمانی

Recent trust research in the information systems (IS) field has described trust as a primary predictor of technology usage and a fundamental construct for understanding user perceptions of technology. Initial trust formation is particularly relevant in an IS context, as users must overcome perceptions of risk and uncertainty before using a novel technology. With initial trust in a more complex, organizational information system, there are a number of external determinants, trusting bases, that may explain trust formation and provide organizations with the needed levers to form or change individuals’ initial trust in technology. In this study, a research model of initial trust formation is developed and includes trusting bases, trusting beliefs, trusting attitude and subjective norm, and trusting intentions. Eight trusting base factors are assessed including personality, cognitive, calculative, and both technology and organizational factors of the institutional base. The model is empirically tested with 443 subjects in the context of initial trust in a national identity system (NID). The proposed model was supported and the results indicate that subjective norm and the cognitive–reputation, calculative, and organizational situational normality base factors significantly influence initial trusting beliefs and other downstream trust constructs. Factors from some of the more commonly investigated bases, personality and technology institutional, did not significantly affect trusting beliefs. The findings have strategic implications for agencies implementing e-government systems and organizational information systems in general.

Recent information systems research indicates that trust plays an important role in helping users overcome perceptions of risk and uncertainty in the use and acceptance of new technology (Gefen et al., 2003 and Pavlou and Gefen, 2004). Numerous studies have investigated the influence of trust on perceptions and/or use of technology in various contexts including eCommerce (Jarvenpaa and Todd, 1997, McKnight et al., 2002a and Pavlou, 2003), online marketplaces (Pavlou and Gefen, 2004), and recommendation agents (Komiak and Benbasat, 2006). This research suggests that understanding how initial trust toward an information system is formed is essential for promoting the adoption of a new system. Research on technology adoption and innovation has noted that the determinants of initial adoption differ from the determinants of continued use (Agarwal and Prasad, 1997, Chin and Marcolin, 2001 and Karahanna et al., 1999). The adoption literature suggests a strong relationship between users’ pre-implementation expectations of a new system and post-implementation experiences with the system (Ginzberg, 1981 and Staples et al., 2002). From a diffusion of innovations perspective, users initially seek or are provided with information about an innovation which forms their attitude toward the innovation and subsequent adoption decision (Rogers, 1995). The decision to readily adopt a new technology is influenced by users’ initial perceptions of the technology characteristics (Moore and Benbasat, 1991 and Rogers, 1983), and users “will be less likely to experiment with new technologies if they perceive a significant risk associated with such exploration…” (Agarwal and Prasad, 1997). This initial exposure to a system is also said to be a time of importance as this is when adaptation of the system and user learning is most likely to occur (Tyre and Orlikowski, 1994 and Weick, 1990). In the context of organizational systems, adoption research has also found that in the early stages of technology adoption, indirect information such as perceptions of the organizational sponsor, organizational structure, and social influence will have the greatest influence on adoption decisions (Gallivan, 2001). This perspective on initial adoption of technology aligns with research on initial trust. While trust is a dynamic concept that develops over time, researchers have noted the importance of studying initial trust, especially in cases of novel technology where users must overcome perceptions of risk and uncertainty before using the technology (McKnight et al., 2002b and Wang and Benbasat, 2005). Trust research in organizational contexts has demonstrated that individuals form trusting beliefs prior to having first-hand experience with another party and that high levels of initial trust exist (Berg et al., 1995 and Kramer, 1994) and thus may be cultivated. The determinants of initial trust are believed to differ from the determinants of trust with a more familiar party (Gefen, 2004 and McKnight et al., 2002a), as individuals do not have direct experiences to draw upon in forming initial trusting beliefs. Research on initial trust is also relevant to later trust with a familiar party as all trusting relationships have a starting point (a beginning), and influential events often occur at the beginning of a relationship that may affect the continuance and/or later phases of a relationship (McKnight and Chervany, 2006). In the context of IS trust, organizations have an opportunity to create a positive, first impression of a new system and generate high levels of trust with users before they interact with the system. In order for organizations to take advantage of the opportunity to create a trustworthy first impression of a new system and encourage adoption, the relevant determinants of trusting behaviors in the context of initial IS use need to be studied. McKnight et al. (2002a) developed and tested a model of initial trust in an IS context based on the theory of reasoned action (TRA) (Ajzen and Fishbein, 1980). Their model included two trusting bases as external determinants (personality and institutional), trusting beliefs, and trusting intentions. This model provides a parsimonious, essential starting point for studies of initial IS trust. Ajzen and Fishbein (1980) suggest that to better understand a behavior or the intention to perform a behavior, it is important to develop a thorough understanding of the underlying external variables in the behavioral context. The organizational trust literature suggests several theoretically-grounded trusting bases, such as cognitive, calculative, institutional, and personality, that may influence initial trusting beliefs (Mayer et al., 1995 and Rousseau et al., 1998). Studies of IS-related trust have investigated only a subset of these trusting bases, and the object of trust in these studies (the trustee) has ranged from virtual team members to web vendors and online sellers. The operationalization of trusting beliefs and intentions has also varied across IS trust studies, and recent empirical research provides evidence that more parsimonious, one-dimensional representations may produce erroneous results (Serva et al., 2005). While many advances have been made in IS-related trust, there is still much to be learned about the determinants of initial trust in information systems. The objective of this research is to provide insight on how to build trust in new technology prior to actual use of the technology. An empirical investigation on the formation of initial trust toward an organizational information system is conducted using a research model of trusting bases, trusting beliefs, trusting attitudes and subjective norm, and trusting intentions. Eight trusting base factors are assessed including personality, cognitive, calculative, and both technology and organizational factors of the institutional base. The model is tested with survey data collected in the context of a national identity (NID) system; the findings provide insights to organizations implementing or updating information systems. An NID system was selected as the context for the study due to its organizational scale, complexity, and the lack of an NID system within the United States (U.S.) where the study took place. An NID system is a large-scale, public information system used at the governmental-level to facilitate public services, enhance national security, and guard against illegal immigrants (Clement et al., 2001). Many European and Asian countries already utilize such systems and advanced user features such as online access to personal data, other government services, and electronic funds transfer with NID cards are often provided (European eGovernment Services, 2005, Finnish Population RC, 2005 and Hong-Kong, 2004). An NID system has been advocated in the U.S. for several decades, but opposition to such a system has remained strong. Campaigns against NID systems have capitalized on citizens’ fear and distrust of such systems (Burns, 2001 and Dority, 2002) and have played a part in preventing the adoption of NID systems. In the specific context of NID systems, an understanding of the external factors that influence citizens’ initial trust in these systems will be of assistance to government agencies implementing e-government systems. In a general context, an understanding of these determinants of initial trust will be insightful for organizations implementing new, complex systems. The paper is organized as follows: the theoretical foundation, research model, and hypotheses are first presented, followed by a discussion of the research methodology. The data analysis is then described, and a discussion of the results is provided. Lastly, the conclusion, limitations, and directions for future research are discussed.

This study proposed and tested a model of trusting bases with initial trust formation in the context of NID systems. The empirical investigation of the trusting bases and social influence as well as the relative effects on trusting beliefs, attitudes, and intentions enhances our understanding of initial trust formation. The findings provide practical implications to government agencies implementing new systems, and the initial trust model can be generalized to organizations implementing other types of information systems. The specific findings related to cognitive and calculative bases of trust have significant academic and practical implications. Past research on trust in information systems has focused less attention on cognitive and calculative bases and has not evaluated these bases in a more comprehensive model that includes other trusting bases and complete measures of trusting beliefs, attitudes, subjective norm, and trusting intentions. From a practical perspective, the findings related to cognitive and calculative bases provide governments with actionable recommendations for improving trusting beliefs in an NID system. Cognitive-based beliefs can be enhanced through reputation-building mechanisms, while calculative-based beliefs can be enhanced by establishing and publicizing policies and procedures that make untrustworthy behavior with an NID system very costly. Past studies of trust in information systems have focused more on personality and institutional bases, and findings related to these bases are less actionable, as personality-based beliefs are typically stable over the lifetime of a trustor, and the institutional context is larger in scope and more difficult to control or manipulate. Limitations of the study may include construct operationalization. In this study, the trusting base factors were operationalized in a general manner. Specific system and product characteristics were not included in the model or measured, as a U.S. NID system does not yet exist. Technology structural assurance and situational normality, two factors of the institutional base, were operationalized in the context of general information systems, instead of NID systems, as most subjects in the study did not have prior experience or knowledge of NID systems. A more complete, specific set of external determinants could explain more variance in trusting beliefs and thus provide more insight to institutions implementing organizational systems. A limitation in the study design is the use of student subjects. While the students represent valid, potential users of an NID system, broader samples could be used in future studies to better address population differences such as age, occupation and computer experience. Another limitation in the design includes the use of a field survey rather than an experiment. While field surveys offer external validity, internal validity is limited, as a controlled manipulation of the causal factors is not conducted (Blackburn, 1987 and Zmud et al., 1989). Future research is needed with the actual use of an NID system to investigate the reciprocal relationships that can exist among trust constructs after initial trust is formed and a long-term trusting relationship is established (Gilliland and Bello, 2002 and Verhoef et al., 2002). Cultural issues should also be investigated. While subjective norm may subsume some cultural issues, such as an increased concern for privacy in the U.S., future research could investigate cultural differences by comparing initial trust across cultures. Lastly, confirmatory studies of the research model and the relative influence of the trusting bases should be conducted in other contexts where new, complex forms of information technologies are being deployed. Different system contexts could influence the relative importance of the various trusting bases as determinants of trusting beliefs.