I am quite new to statistics. I have emotion response and preference response from normal hearing and hearing impaired people for both 28 piano and drum clips. I have two factors tempo of the songs and mode of the song (major or minor). I want to look at both the fixed effects of tempo and mode and the interaction effects on each of these groups.
I expect somthing like this (which I dont get using Fit Model):
Emotion rating = a*Mode + b*Tempo + c*Tempo*Mode
Preference rating = a*Mode+ b*Tempo + c*Tempo*Mode
Emotion and preference - continuous variable, Tempo - continuous, Mode - Categorical (Major or Minor)
The way I do this is, Analyse ----> Fit Model ----> I choose Tempo and mode then full factorial under Macros to get main effects and interaction effects. For emotion rating (for drums), And for the y-column I choose both normal and hearing impaired response for emotion for drums. I do the same for piano response for both the groups. I am not looking at the variablity within the subjects in group but the effect of tempo and mode for drums and piano on emotion and preference rating for the clips. Also, I am looking at Random effects, it should have been looked into but for technical reasons its eliminated.
I have already tried this approach but I dont know if this is the right way to look the data when there are two different groups and and all other variables are same.
