RESULTS AND ANALYSES
Linear structure model analysis
The CFA is an analytical approach opposite to the Exploratory Factor Analysis (EFA). This study conducted a CFA of three unobservable/implicit variables (i.e., internal-service quality, employee’s job satisfaction and organizational performance). Consisting of the Structural Model and Measurement Model, the SEM provides an effective solution to the cause-effect relation between implicit/latent variables. Besides, the models verified in this study have three parts: (1) verifying the goodness-of-fit of the Measurement Model; (2) verifying the goodness-of-fit of the Structural Model and (3) verifying the overall model’s goodness-of-fit to make sure it conforms to the goodness-of-fit indices. That is, the goodness-of-fit of the overall SEM was judged with related goodness-of-fit indices. (Diamantopoulos & Siguaw, 2000)
Analyzing fit of the Measurement Model
The factor loading of latent/implicit variables and manifest/ explicit variables mainly measures the intensity of linear correlation between explicit and implicit variables. A factor loading close to 1 indicates the explicit variable is relatively capable of measuring the implicit one. In this study, all explicit variables’ factor loading are between 0.7 and