Maximum likelihood was used for all parameter estimation with Amos 16. First confirmatory factor analysis (CFA) is conducted to evaluate the model used to measure the modeled constructs. CFA enables testing of the reliability, convergent validity and discrimi- nate validity of the measurement model. The reliability and internal validity of the measurement model is examined by calculating the composite reliability (CR) and average variance extracted (AVE). Table 2 shows that all the constructs have acceptable composite reliability coefficients, since they exceed 0.6 (Bagozzi and Yi, 1988; Hair et al., 1998). The AVE of each measure accounts for more than 50% of the variance, as suggested by Bagozzi and Yi (1988), and indi- cates that the variance captured by the construct exceeds that due to the measurement error (Fornell and Larcker, 1981). Therefore, the measurement model has adequate internal validity.