EFA and CFA Measure purification was performed on each scale using factor analysis (exploratory and confirmatory) and reliability measures. Somemeasuresweredroppedafterfactoranalysesindicated cross-loading. The results of additional EFAs indicated a rotated seven-factor solution, with the items corresponding to the factors for each construct in the model (explaining a total variance of 72.15 per cent, with the Kaiser-Meyer-Olkin (KMO) test equal to 0.864, significant at p 0.001)[4]. We then conducted a confirmatory factor analysis (CFA) using Lisrel 8.8. As shown in Table III, composite reliabilities (rho, ) ranged from 0.77 to 0.92, well above Bagozzi and Yi’s