I have a set of answers to the shortened version of the IPPA questionairre (N ~ 200). Each person is asked 12 questions twice (in a single questionnaire), ascertaining the relationship with the mother and father consecutively.

I would like to assess the differences in answers between the answers for the father and mother. I have devised sub-scales and calculated means. But what is the right way to test/compare these values? Paired-sample T-tests? RM-Anova?

I will do a correlation test also, because that information is also important to me.

Any thoughts?