I don't quite understand your design, but in general, if you are using multiple tests to support what might be the same research question (e.g., if you can say "yes, my theory was right" if even just one of these tests comes out significant) then you need to correct for multiple comparisons based on the number of tests you did.
I am interested in the correlation effect of a sequence of conditions. If a strategy to solve a task holds the "same" across the sequence, so results are correlated across the conditions. would it be ok to correct it "fdr"?
FDR (and other multiple hypothesis correction methods) are for situations where you are testing alternative hypothesis (is A correct OR is B correct OR is C correct).
If you have a situation where you need A AND B AND C to be correct then FDR is not required and would be detrimental to apply. This is because you have one hypothesis, even if it is made of of several conditions each tested separately.
Your follow up comment sounds like you need all conditions to pass, if that is the case it sounds like FDR is inappropriate.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix