Organizations invest in technology to help achieve strategic goals and to disseminate knowledge in order to enhance employee productivity. This study draws upon social cognitive theory as the theoretical basis for an explanation of the limited acceptance of information systems (IS), which undermines organizations' efforts to achieve greater productivity. This empirical study investigated the use of IS in the workplace by examining the antecedents of computer self-efficacy (CSE). Among such antecedents, this study examined the effect on CSE of employees' knowledge sharing, including collegial and technical support, and encouragement by others such as top management, supervisors, and peers to share knowledge. Additionally, it assessed the differences that the control variable, use of social networking tools (SNT), has on CSE and use of IS in the workplace. This research explored the use of SNT as a way individuals share knowledge and encourage computer abilities in the workplace.

This causal modeling study investigated the human-assisted CSE dimension, which focuses on one's ability to use IS resulting from the support of another individual. This study solicited 755 participants from five different organizations, with 256 responses (33.9% response rate). The Cronbach Alpha analysis results showed that all constructs were highly reliable. The structural equation model analysis showed that some, but not all, of the hypotheses were supported in this model.

This study provided evidence of how the antecedents, knowledge sharing and encouragement by others, affected the constructs of CSE and the use of IS in the workplace. Important contributions of this study include not only the conceptual model itself but also the implication that encouragement by top management, supervisors, and colleagues plays an important part in knowledge-sharing behaviors. Additionally, the study makes a theoretical contribution to the growing body of research on SNT as a way to share knowledge. This study revealed that use of SNT had no significant effect on the constructs measured.