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

Online Travel Product Attributes Factor Analysis 
Factor analysis and reliability analysis were conducted to examine the underlying construct of online 
shopping attribute of travel suppliers. The results of factor analysis, using principal component analysis and 
Varimax rotation procedure showed that three factors are appropriate with eigenvalue of greater than 1 (see Table 
1). The results showed that there are appropriate correlations among the variables. The Bartlett’s test of sphericity 
indicates that sufficient correlations exist among the variables (Approx. chi-square = 1104.752, df = 55, p = .000) 
and the measure of sampling adequacy value exceeded the recommended cut-off of 0.8 (KMO=.839). The decision 
on the number of factors was based on several criteria; factors with eigenvalue greater than 1.0, percentages of 
variance explained, and Scree plot. All criteria indicated a three-factor solution was appropriate with a explained 
variance 62.213%. Each factor’s loadings were greater than 0.5, which were then named “Perceived value,” 
“Product information,” and “Site properties.” Construct Cronbach’s alpha values were greater than 0.60, indicating 
each construct is statistically meaningful to measure each construct. Frequent membership loyalty program was 
deleted due to low communality with .381 and anti-image value. Total 11 attributes were analyzed for the factor 
analysis.