رفتار به اشتراک گذاری دانش در اجتماعات مجازی: رابطه بین اعتماد، خوداثربخشی، انتظارات و نتایج

There has been a growing interest in examining the factors that support or hinder one's knowledge sharing behavior in the virtual communities. However, still very few studies examined them from both personal and environmental perspectives. In order to explore the knowledge sharing behaviors within the virtual communities of professional societies, this study proposed a social cognitive theory (SCT)-based model that includes knowledge sharing self-efficacy and outcome expectations for personal influences, and multi-dimensional trusts for environmental influences. The proposed research model was then evaluated with structural equation modeling, and confirmatory factor analysis was also applied to test if the empirical data conform to the proposed model.

The Internet enables knowledge exchange in ways such as online group meetings that were not possible before. It also gave rise to professional virtual communities (VCs) that enable knowledge sharing without ever meeting the participants. Today, more and more individuals participate in VCs to acquire knowledge to resolve problems at work. Many organizations have also recognized the VCs as a valuable system that holds the key to knowledge management and have begun to support the development and growth of VCs to meet their business objectives. A VC is a cyberspace supported by information technology. It is centered upon the communications and interactions of participants to generate specific domain knowledge that enables the participants to perform common functions and to learn from, contribute to, and collectively build upon that knowledge (Lee et al., 2003). While VCs bring people together “virtually” from all over the world, knowledge sharing among them has not lived up to expectation. Davenport and Prusak (1998) argue that sharing knowledge is often unnatural because people think their knowledge is valuable and important; hoarding knowledge and being suspicious upon knowledge from others are the natural tendency. Furthermore, Pfeffer and Sutton (1999) find that knowledge management in many organizations only emphasize on technology, particularly information technology. Finally, Dixon (2000) point out that “build it and they will come” and “technology can replace face-to-face interaction” are the myths of knowledge sharing. Clearly, the biggest challenge in fostering VCs is the willingness to share knowledge with other members. In this respect, two issues are involved: personal cognition and social influence; personal cognition is based on self-efficacy and outcome expectations and social influence is based on trust. Identifying the personal cognition (Bartol and Srivastava, 2002; Bock and Kim, 2002; Bock et al., 2005; Kankanhalli et al., 2005) and the relationships among members’ underlying knowledge sharing behavior in VCs (Wasko and Faraj, 2005; Bock et al., 2005) would help both academics and practitioners gain insights of how to stimulate knowledge sharing in VCs. To this end, a foundation for investigating this matter is the social cognitive theory (SCT). SCT (Bandura, 1986 and Bandura, 1997) has been widely applied in the information systems literature with demonstrated validity. This theory states that an individual will take an action that has personal cognition in a social environment. Furthermore, a person's cognition to act in a certain way has two basic determinants: self-efficacy and outcome expectation. Self-efficacy, or the belief in one's capabilities to organize and execute courses of actions required to manage prospective situations (Bandura, 1997), is a potentially important factor influencing the decision to share knowledge (e.g., Bock and Kim, 2002; Hsu et al., forthcoming; Kankanhalli et al., 2005). Outcome expectations that are related to reward systems (Bartol and Srivastava, 2002) are also important factors influencing the decision to share knowledge. According to the economic exchange theory, individuals will behave by rational self-interest, thus, knowledge sharing will occur when its outcomes exceed its costs or are as expected (Constant et al., 1994). This is why practitioners emphasize incentive systems for successful knowledge management. On the one hand, outcome expectations imply that, if members of VCs believe that they would receive extrinsic benefits such as monetary rewards, promotion, or educational opportunity from their knowledge sharing, then they would develop a more positive attitude toward knowledge sharing (Bock and Kim, 2002; Kankanhalli et al., 2005). On the other hand, if members believe that they would receive intrinsic benefits such as self-satisfaction, social recognition, or power, then they would also have pleasure in knowledge sharing (Kankanhalli et al., 2005). Trust, an implicit set of beliefs that the other party will behave in a dependent manner (Gefen et al., 2003; Kumar et al., 1995) and will not take advantage of the situation (Gefen et al., 2003), has been recognized as an important factor affecting knowledge sharing (Ridings et al., 2002). Knowledge sharing is a motivation for members to use VCs (Ridings et al., 2002; Wasko and Faraj, 2000). In the context of VCs, members voluntarily contribute their knowledge without receiving monetary rewards (Lee and Cole, 2003). Unlike traditional organizations, membership in VCs is open (Lee and Cole, 2003) and members of VCs are formed by common interests and practices (Ba, 2001; Ridings et al., 2002) without shared norms and routines to serve as linkage between members and VCs. Thus, the relationship between members and VCs are more fragile than that of traditional organizations. Moreover, most members in VCs are relatively invisible, and most VCs do not provide guarantees that others will behave as they are expected to (Ridings et al., 2002). The lack of face-to-face communication and legal guarantees makes it harder for members of VCs to share their knowledge. Hence, trust is important in VCs, because it could create a necessary atmosphere that makes interaction with others more open (Bulter and Cantrell, 1994; Ridings et al., 2002) and rules out the undesired and opportunistic behaviors (Luhmann, 1979; Ridings et al., 2002). In these perspectives, trust is a crucial factor to sustain the continuity of VCs (Ridings et al., 2002). Overall, the purpose of this study is to identify the antecedents that support or hinder an individual's knowledge sharing behavior by applying SCT-based model from both social environment and personal cognition aspects. Especially, this study aims to explore the nature of trust and divides it into three constructs—economy-based, information-based and identification-based trust—to examine their impacts on an individual's knowledge sharing self-efficacy and behavior. The knowledge sharing self-efficacy is served as a behavioral control variable to deal with situations in which people face the challenge of exchanging knowledge among individuals in cyberspace. This paper is organized as follows: first, the SCT is described as a theoretical background to link self-efficacy, outcome expectation, and trust, and a research model is proposed. Then, research methodology and data analysis are discussed. Finally, the conclusion and limitations are presented.

5.1. Findings and implications SCT has been widely used and validated for human behavior in numerous contexts, but it still has not been applied to knowledge sharing. This study aims to shed light on the knowledge sharing behavior in VCs and is the first study that applied SCT-based model to investigate the determinants of knowledge sharing behavior. This study validates the proposed research model and the findings presented herein respond to the research questions. First of all, the results indicate that self-efficacy has both direct and indirect effects on knowledge sharing behavior, implying that self-efficacy plays a critical role in guiding individuals’ behavior. From the practitioners’ standpoint, management of professional VCs should provide some strategies (e.g., online training programs, support mechanism and many others) to increase members’ self-efficacy so that people would believe they will be able to share their knowledge in the professional VCs. Second, our findings show that personal outcome expectations have significant influence on knowledge sharing behavior. This finding is consistent with Bock and Kim's study (2002). They argue that if individuals believe they could improve relationships with others by offering their knowledge, they would develop a more positive attitude toward knowledge sharing. The greater the anticipated reciprocal relationships are, the more favorable the attitude toward knowledge sharing will be (Bock et al., 2005). Therefore, members who think knowledge sharing would increase the scope and depth of associations among VCs members tend to share knowledge with others. Moreover, it may be necessary for managers of professional VCs to offer reward mechanisms (e.g., award of best knowledge contributor, ranking of knowledge sharing) to raise individuals’ positive personal expectations. Once individuals in the professional VCs perceive the future reward, they would be likely to contribute knowledge with others. Third, our findings suggest that community-related outcome expectations have no significant influence on knowledge sharing behavior. A possible explanation for this finding is that VCs comprise of people with similar interests or needs to carry out knowledge-based communication in voluntary settings. Unlike other types of organization (such as firms, government units), such virtual organizations do not have formal and power rules, routines and procedures to guide the users’ knowledge sharing behaviors as asserted in the institutional theory (Purvis et al., 2001) and they also seldom offer incentive systems for knowledge contributors. Members in the VCs expect to share the knowledge they are interested to reach their personal goals (e.g., praise, promotions, image, social status). Hence, members are likely to share their knowledge when positive personal outcome expectations can be realized. Fourth, our findings reveal that economy-based trust and information-based trust has to be established first, and then develop identification-based trust. Only by forming these kinds of trusts, mutual trust will be formed. Trust is not a single or unidimensional concept and develops gradually as the parties move from one stage to another in an organization context (Boon and Holmes, 1991; Lander et al., 2004; Panteli and Sockalingam, 2005). Several researchers (Ba, 2001; Luo and Najdawi, 2004) also argue that trust building stages exist in online environment. Furthermore, to the best of our knowledge, this study is one of the earliest empirical studies that use this standpoint to examine individuals’ knowledge sharing behavior and provide insight into the trust building processes in professional communities. From a practical perspective, managers should assist members to move from economy-based trust to information-based trust. When new members participate in VCs, they may behave according to the expected rewards and costs incurred by their behavior. Therefore, economy-based trust entice member into joining the VCs. Members could reduce risk and uncertainty by developing information-based trust. To remove barriers, managers should facilitate self-regulating policies, source disclosure, third party seals, and establish brands of VCs. Managers should make sure members’ personal data is secure, let them know this place is well managed, and encourage them to interact with each other. Finally, this study finds that identification-based trust plays a critical role in knowledge sharing behavior. In the VCs, members have similar interests, goals and objectives, and believe they will do similar things in a similar situation. Identification-based trust would have the strongest emotional component, and much of its development centers around common interests, goals, values and principles (Lewicki and Stevenson, 1997). Therefore, VCs should be conducted in a cordial atmosphere where members could establish familiarity through frequent interactions and easy communication, and then they will move from information-based trust to identification-based trust. Eventually, managers should encourage the members to broaden their professional abilities and then contribute their knowledge to every member. Members will have more willingness to share their knowledge with other members, when they believe that they will strengthen the ties between existing members, expand the scope of their association with other members, and draw smooth cooperation from outstanding members in the future. Thus, successful knowledge sharing is a result of identification-based trust among members. 5.2. Limitations and suggestions for further research Even though this study has offered valuable insights into knowledge sharing, it has some limitations as most field surveys suffered. First, the study did not take community type into consideration, even though a variety of professional communities were explored in this study. It means the different properties of different professionals might be overlooked. Hence, it may limit the generalizability of this study to overall knowledge sharing behavior in the VCs. Second, the study did not distinguish the members according to the time they spent. Since trust increases over time, we cannot get an insight into whether members in different stages perform different knowledge sharing behavior. This study investigates the staged development of trust and the direct impact of identification-based trust on knowledge sharing behavior. As to the impact of economy-based trust and information-based trust on knowledge sharing behavior, we leave them to future studies due to space limitation. Third, this study examined the effect of environment and personal factors on knowledge sharing behavior and the relationship between these factors. Actually, according to the SCT (Bandura, 1986 and Bandura, 1997), determinants influence each other bidirectionally, that is, behavior in VCs is affected by environment and personal factors, which in turn are influenced by behavior (Compeau and Higgins, 1995a). Hence, we hope future study may conduct a longitudinal survey to examine the interactive relationship among the three reciprocal determinants. Fourth, this study also suggests the concept of multistage trust. Therefore, in the future, it would be interesting to expand trust as multilevel constructs and examine their relationships and further examine how they influence knowledge-sharing behavior. Fifth, most researchers agree that common method variance (i.e., variance that is attributable to the measurement method rather than to the constructs the measures represent) is a potential problem in behavioral research. Method biases are a problem because they are one of the main sources of measurement error. Measurement error threatens the validity of the conclusions about the relationships between measures and is widely recognized to have both a random and a systematic component (Bagozzi and Yi, 1991; Podsakoff et al., 2003). Self-selection issue is the common problem of the questionnaire survey process, especially a methodology concern. It may be possible in this study that responses come from individuals with a predilection to share, and their response may not consider what influences one not to share frequently. Finally, this study only focused on the most important factor (i.e. trust) of SCT, other relational constructs in the environmental dimensions that can affect knowledge sharing behavior should also be examined, such as subjective norms (Ryu et al., 2003) and mutual influence (Nelson and Cooprider, 1996). Social capital theory has been used to discuss the relational construct as a broad view in exchanging and combining intellectual capital—including structural, cognitive and relational dimensions (Nahapiet and Ghoshal, 1998; Tsai and Ghoshal, 1998). As a result, it would be interesting for further research to extend the present model with the social capital theory to investigate the relationships between these dimensions and knowledge sharing behavior.