Partial least squares (PLS) were used in this study to assess the measurement model within the context of its theoretical mediated model, making it superior to multiple regression, especially when using new scales. PLS employs component-based estimation (Lohm€oller, 1989), thereby maximizing the variance explained in the dependent variable.It does not require the data to have multivariate normality (Gefen et al., 2000) and it can handle formative constructs (Chin, 1998).Furthermore, it imposes fewer estimates for path coefficients of the structural model.