Exploring a Broader View of Technology Acceptance
The primary aim of this dissertation is to establish the generalizability of the scale items used to measure 5 psychological acceptance constructs proposed by Schwarz and Chin (2007). While an initial test of validity and reliability was established by Schwarz (2003) using covariance-based structural equation modeling, a stronger test was performed to establish the generalizability of the items through a series of multigroup invariance tests. Having used 3 new independent data sets, we present the results of the combinatorial analyses of 3 pairwise comparisons of the data sets as well as a test comparing all 3 data sets simultaneously. Both confirmatory factor models and structural models were applied to examine whether item measures are identically reliable and whether the relationships among these 5 constructs also remain the same. Structurally, two models incorporating these 5 constructs were applied to predict an overall general acceptance construct and the construct of infusion. While the nomological relationships among these acceptance constructs varied as expected, the correlations and item loadings remain invariant. Therefore, the results answer the questions: (1) Can the acceptance constructs proposed by Schwarz and Chin be captured by reliable and accurate measures? (2) Are these constructs distinct from one another? and (3) Do they act similarly in different contexts? Finally, to provide a platform for more research on workplace outcomes, this research explores the notion of technology infusion, an important form of usage. Given that the 5 psychological acceptance constructs have predictive value toward infusion, we establish a means for further study of the concept.