 |
 |
You are committing a statistically fallacy here, by trying to say A and B are equal. You cannot demonstrate that with that contrast. If p < 0.05 we reject the null hypothesis that A and B are EQUAL. If p > 0.05, all we can say is that we are not certain that any difference between A and B is a true population difference for our probability level.
You cannot really show two groups are equal based on a traditional t-test. In fact that is very tricky to do. The best you can do with this approach is show they are not STATISTICALLY different at a given threshold.
Colin Hawco, PhD
Neuranalysis Consulting
Neuroimaging analysis and consultation
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Qasim Bukhari
Sent: March-15-16 12:30 PM
To: [log in to unmask]
Subject: [FSL] design matrix for testing A=B vs A!=B
Dear FSL experts
I was running a study, and just wanted to make sure that my design matrix is correct
I want to test whether my group A is equal to group B or not(both having 8 subjects). Null hypothesis is that it is NOT equal, and I want to prove it false. (that means if A and B are equal I m happy)
Should these be my mat. file
1 1 0 0 0 0 0 0 0
1 0 1 0 0 0 0 0 0
1 0 0 1 0 0 0 0 0
1 0 0 0 1 0 0 0 0
1 0 0 0 0 1 0 0 0
1 0 0 0 0 0 1 0 0
1 0 0 0 0 0 0 1 0
1 0 0 0 0 0 0 0 1
-1 1 0 0 0 0 0 0 0
-1 0 1 0 0 0 0 0 0
-1 0 0 1 0 0 0 0 0
-1 0 0 0 1 0 0 0 0
-1 0 0 0 0 1 0 0 0
-1 0 0 0 0 0 1 0 0
-1 0 0 0 0 0 0 1 0
-1 0 0 0 0 0 0 0 1
and con file
1 0 0 0 0 0 0 0 0
-1 0 0 0 0 0 0 0 0
Would the design matrices remain the same if I reverse my null hypothesis ? That means to make the null hypothesis that both groups are same (optimal result that we disprove it). What would be the change in the design matrix then?
I think I m doing something wrong, if anyone can help in this, I ll be grateful ?
Thanks and kind regards,
Qasim