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

 
Complaint Intention. 
First, standard multiple regression was conducted to allow to determine which attribute 
predict the traveler’s complaining intention. The overall regression equation (
R
2
=.199; 
F
=24.418, 
p
<.05) was 
statistically significant. Only two of the independent variables contributed significantly to the prediction of traveler’s 
complaining intention (i.e., Perceived value (
β
=.405, p=.000) and Site properties (
β
=.126, p=.033). Perceived value 
was a stronger predictor of complaining intention than was Site properties. At step 2, the loyalty program use added 
to the prediction of complaining intention by online travel product attributes. When loyalty program was added into 
the online travel product, 
R
2 
was increased from .199 to .226 (.027); thus, 2.7% of the total variance is uniquely 
accounted for by loyalty program use. The difference test between two models was statistically significant because 
F-value (2.503) of 
Δ
R
2 
= .027 is greater than critical value of 2.503 at the level of .05. Furthermore, the two 
interaction effect (i.e., perceived value*loyalty program use, product information*loyalty program use). Thus, the 
significant increment to the variance accounted for by the prediction model affirms that the loyalty program use has 
a moderating effect to online travel product toward tourists’ complaining intentions.