Hello,
I’m sorry I’m asking a question related to what others have previously raised but after searching the main wiki and this board I was not successful in finding the specific solution.
I have a within-subjects design: each subject completes 4 conditions (2 factors (A, B), each factor has two levels (1,2)). For each of the 4 conditions, each subject contributes 2 observations (scan1, scan2).
A 2 x 2 repeated measures within-subjects ANOVA is the test I seek.
I now have one subject (so 4 conditions x 2 scans = 8) but will soon expand to have several subjects (so single “group” only, within-subjects analysis).
I gather expanding a 2 x 2 ANOVA model to account for my 2 non-independent repeated measures is not trivial, but I have attempted modeling the 2 observations per condition in the vein of paired t-tests.
My file order is (for my single subject)
1. Scan1_ConditionA1B1
2. Scan2_ConditionA1B1
3. Scan1_ConditionA1B2
4. Scan2_ConditionA1B2
5. Scan1_ConditionA2B1
6. Scan2_ConditionA2B1
7. Scan1_ConditionA2B2
8. Scan2_ConditionA2B2
In addition to the 4 EVs (A1B1, A1B2, A2B1, A2B2), I’ve included an EV to account for scan number (scan1, 2) but I'm afraid I didn't model it correctly.
I feel the hybrid model and contrasts I’m attaching are not valid, and am hoping for feedback and corrections.
Thank you kindly,
Yuliya
Here is the design http://imgur.com/uO0uNUX
|