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

 
2007 Annual International CHRIE Conference & Exposition 
185
Convergent and Discriminant Validity
Convergent validity exists when a researcher’s proposed scale items correlate with the same construct. The 
standardized loadings and the squared multiple correlation (SMC) for the measurement items and the constructs 
were examined as an evidence of convergent validity (Bollen, 1989). Significant factor loadings for a specified 
construct suggest confirmation of convergent validity, indicating that indicators for a given construct should be at 
least moderately correlated among themselves. All items loaded significantly on their specified construct with the 
lowest critical ratio (t-value) being 12.07 (p < .01), which indicates that the specified indicators are sufficient in their 
representation of the constructs (Hair, Anderson, Tatham, & Black, 1998).
To evaluate the discriminant validity of the constructs, a confidence interval test approach suggested by 
Anderson and Gerbing (1988) was adopted. A confidence interval of plus or minus 2 standard errors was first 
calculated around the correlation coefficients between the constructs and tested if any interval includes 1.0. In this 
test, none of the confidence intervals included 1.0; thus, the correlation between the constructs was not strong and 
discriminant validity is confirmed (Bagozzi, 1980; Anderson & Gerbing, 1988; Hatcher, 1994). In summary, these 
results demonstrate that the proposed measurement model is appropriate for further analysis.