A Web-based questionnaire was developed to include multi-item scales measuring the constructs in the framework. Some of the items were adapted from previous research projects and modified to fit the context of this study. New items were developed through a thorough literature review of the constructs. An electronic mail message that explained the objectives of this study and directed users' web browsers to the survey website was distributed to 1,865 Internet users to collect data. After the initial screening for usability and reliability, 253 responses were found to be complete and usable.
Confirmatory Factor Analysis (CFA) was used to assess the validity of the measurement model. Based on both the statistical results and the understanding of the subject, the final measurement model was re-specified to include 43 items out of the 64 items in the initial questionnaire. The comparative fit index (CFI) for the final measurement model was found to be 0.90, indicating an adequate fit.
Structure Equation Modeling (SEM) technique was then used to evaluate the causal structure of the proposed framework. Model generating (MG) strategy was applied to modify the proposed model until it was both theoretically meaningful and statistically well fitting. The initial statistical results indicated that some revisions were needed. After re-specification, the goodness-of-fit statistics indicated that there is a good fit between the data and the proposed framework (CFI = 0.91). A schematic representation of the final structural model for the Model of Consumer Acceptance of Virtual Stores that includes the standardized path coefficients is displayed in Figure 2.