Marketing Research (SPSS)
bingjieyu9
_____08dDVsa2x2ExperimentalDesign-QualityandEconomy3.pptx8. A 2 x 2 Experimental Design: - Quality and Economy (x1 and x2 as independent variables)
Dr. Boonghee Yoo
RMI Distinguished Professor in Business and
Professor of Marketing & International Business
Make changes on the names, labels, and measure on the variable view.
Check the measure.
Have the same keys between “Name” and “Label.”
Run factor analysis for ys (dependent variables).
Select “Principal axis factoring” from “Extraction.”
The two-factor solution seems the best as (1) they are over one eigenvalue each and (2) the variance explained for is over 60%.
The new eigenvalues after the rotation.
The rotated factor matrix is clear.
But note that y3 and y1 are collapsed into one factor.
If not you should rerun factor analysis after removing the most problematic item one at a time. Repeat this procedure until the rotated factor pattern has (1) no cross-loading, (2) no weak factor loading (< 0.5), and (3) an adequate number of items (not more than 5 items per factor).
If a clear factor pattern is obtained, name the factors.
Attitude and purchase intention (y3 and y1)
Boycotting intention (y2)
Compute the reliability of the items of each factor
Make sure all responses were used.
Cronbach’s a (= Reliability a) must be greater than 0.70. Then, you can create the composite variable out of the member items.
Means and STDs must be similar among the items.
No a here should be greater than Cronbach’s a. If not, you should delete such item(s) to increase a.
Create the composite variable for each factor.
BI = mean (y2_1,y2_2,y2_3)
“PI” will be added to the data.
Go to the Variable View and change its “Name” and “Label.”
