Construct validity was evaluated by convergent validity and discriminant validity. According
to Henseler et al. (2009) both validities can be confirmed if the indicators fit these
conditions: (a) a factor loading has a higher correlation with another latent variable than
with its respective latent variable; (b) the average variance extracted (AVE) is higher than
0.5; (c) the composite reliability (CR) must be greater than 0.6; and (d) the square root
of each construct’s AVE is larger than its correlations with other constructs. All indicator
factor loadings were well loaded on their constructs; their own loadings were much higher
than the cross-loadings with other constructs (see Table 1). The reliability was evaluated
using Cronbach’s alpha and the values were all above 0.70 (see Table 2), which indicates a
reliable measurement instrument. The AVE values were all above 0.50 and CR values were
all above 0.83 (see Table 2). The square roots of AVE values were greater than the correlations
between any pairs of constructs located in their horizontal lines and vertical columns,
proving that the AVE values for the seven constructs were higher than the variance due to
measurement error. Thus, each construct was verified as distinct and the test of discriminant
validity was successful.