Hi
I've a question about the sample size for randomise. I understand that accuracy could be a problem when the sample size is low. Usually sample size of 50-100 subjects is recommened. If we only have 20 subjects (10 subjects for each group) and no nuisance regressors in the model, will the results of randomise be vailided?
Is there a way to test the sample size for randomise? From the fsl website, it says "The accuracy is easily assessed by generating random noise data and fitting it to your design; the uncorrected P-values should be uniformly spread between zero and one; the test will be invalid if there is an excess of small P-values and conservative if there is a deficit of small P-values". Since the non-brain is zero, will it be ok if there is some excess P-values closed to zero?
Thank you,
Pei
|