Dear Michael,
Thanks for your time and help.
Yes, I indeed followed the 1st approach. In the 1st level, I created two contrasts subtracting the two within conditions of each participant:
- condition 2 - condition 1
- condition 1 - condition 2
Then, in the 2nd level factorial design, I input these contrasts into the scan field (2 per subject), and specified the conditions of each one.
Then, I defined 4 factors:
- subjects (n=x),
- group (g1, g2, g3), (between factor)
- gender (M, F) (between factor)
- condition: pointing to the first-level condition1 and condition2
I created 4 main effects (subject, group, gender and condition) and 3 interactions (group X condition; gender X condition; group X gender X condition).
Following this procedure, is it still statistically incorrect to include the group and gender factors (between factors) and the within-repeated factor (condition)?
I know this might be repetitive, but I would really like to be sure if the previous procedure that I used is correct. If not, can you recommend me a guide/article to help me perform this analysis on the SwE toolbox?
Thanks again for you help.
|