Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

 
RESULTS 
Reliability and construct validity were checked on the composite data (414) of IC and RC because the same 
measurements applied across two samples. The Cronbach’s alpha that was used to estimate the reliability of multi-
item scales was listed for each construct in Table 1: professional education (0.86), social networking (0.81), site 
selection (0.71), staff service (0.95), self-image congruence (0.96), brand awareness (0.87), brand satisfaction (0.91), 
UEBV (0.90), brand trust (0.92), and ABL (0.91). All of the alpha coefficients were above the cut-off point of 0.7 
(Nunnally, 1978), indicating an acceptable level of reliability for each construct. Average variance extracted (AVE) 
was also employed to assess convergent and discriminant validity. As shown in Table 1, all AVE exceeded 0.5, 
suggesting convergent validity was satisfied. Also, an AVE for each construct was greater than squared correlation 
coefficients for corresponding inter-constructs, which confirms discriminant validity (Fornell & Larcker, 1981).