Dear Bhavna,
>Sorry to bother you again. I am in the process of writing up my thesis and I
>came across your phrase '"represent the positive and negative (or reverse)
>tails of the interaction" in factorial analysis of our attention -PET study.
>Just for my understanding could you explain what positive and negative tails
>indicate?
Its just the difference between the positive and negative interaction
contrasts specified with weights like
1 -1 -1 1 vs. -1 1 1 -1
It simply means that a one-tailed test of an interaction with the t-statistic
does not test for the negative or reverse contrast (i.e. you have to
test separately for Factor 2 having a positive effect on the response,
due to Factor 1, and for it having a negative effect).
All the very best - Karl
|