Start Date
10-12-2017 12:00 AM
Description
Despite decades of technological and organizational change, our research on technology adoption and acceptance continues to use measures that were developed during the late 1980s and early 1990s. In this paper, we examine one such measure, computer-self-efficacy. We consider the implications of changing technologies and context and propose a new direction for conceptualizing and measuring self-efficacy. We present an updated conceptualization and measurement for a new construct called Technology Self-efficacy. We validated the measure using a survey of 285 technology users. Our analysis shows the scale to be unidimensional and reliable, to exhibit discriminant validity from related constructs, and to possess better predictive validity in relation to outcome expectations and use than the current measure of computer self-efficacy. We conclude with a description of our plans for further validation and an exploration of the implications of our work for self-efficacy theory in information systems, and for construct measurement more broadly.
Recommended Citation
Compeau, Deborah R.; Correia, John; and Thatcher, Jason, "Implications of Technological Progress for the Measurement of Technology Acceptance Variables: The Case of Self-efficacy" (2017). ICIS 2017 Proceedings. 22.
https://aisel.aisnet.org/icis2017/HumanBehavior/Presentations/22
Implications of Technological Progress for the Measurement of Technology Acceptance Variables: The Case of Self-efficacy
Despite decades of technological and organizational change, our research on technology adoption and acceptance continues to use measures that were developed during the late 1980s and early 1990s. In this paper, we examine one such measure, computer-self-efficacy. We consider the implications of changing technologies and context and propose a new direction for conceptualizing and measuring self-efficacy. We present an updated conceptualization and measurement for a new construct called Technology Self-efficacy. We validated the measure using a survey of 285 technology users. Our analysis shows the scale to be unidimensional and reliable, to exhibit discriminant validity from related constructs, and to possess better predictive validity in relation to outcome expectations and use than the current measure of computer self-efficacy. We conclude with a description of our plans for further validation and an exploration of the implications of our work for self-efficacy theory in information systems, and for construct measurement more broadly.