3.3 Data Analysis
Data were analyzed using SmartPLS developed by Ringle, Wende, and Will (2005) known as the PLS-SEM approach. PLS-SEM was used because it has many advantages compared with other first generation techniques. Moreover, PLS as variance-based SEM is preferable because covariance-based SEM assumes that the observed measures have random error variance and they measure specific variance components that are not of theoretical interest and thus, excluded from the measurement model. However, PLS-SEM assumes that the explanation of all observed measure variances are useful.