Dear all,
I have done multiple regression using SPSS for my data, and found that I should have validated the scales first.
So, I went back to check the validity of the scales. One method was using the Confirmatory Factor Analysis (CFA) using LISREL. This analysis CFA showed me that because my data is not normally distributed, I can not run CFA (as the overall Chi square was significant, where it should not) . So, now I know that my data is not normally distributed
My questions are:
1) Having known that my data is not normally distributed, can I still run multiple regression using SPSS?
2) Is it true that even though the overall Chi square was not significant, I still can run CFA to prove the validity of the scales?
3) How can I solve the problem of the data that is not normally distributed?
Your help, advice and suggestions are highly appreciated,
Thank you
Hannah
PhD candidate
---------------------------------
Yahoo! Plus - For a better Internet experience
|