I am conducting hierarchical regression analysis to test whether there
is moderator effect or not on the dependent-independent variables
relationship.
The question is: in SPSS stepwise or enter is the best method?
I came across one reference which said that stepwise is better because
by using this method you are able to decide whether the moderator
effect is pure or quasi (if any). On the other hand, I came across
another reference which emphasised on using “enter” method where the
“inflection point” could only be calculated only when the interaction
term is sig. And also all other related terms (moderator and
independent variable) are found in the equation whether they are sig.
Or not.
I am confused??? What is the best method?
By the way, I tried both and I got very different results, so I do not
think they are the same.
Thank you and looking forward to hearing from you soon
Sarah
__________________________________________________
Yahoo! Plus - For a better Internet experience
http://uk.promotions.yahoo.com/yplus/yoffer.html
|