Thanks Steve for your quick response!
I have a related question that I hope you could answer. We would like to examine Mean Diffusivity, Axial Diffusivity, and Radial Diffusivity in regions where significant FA alteration were found (using the same design.con en design.mat). That would require additional TBSS/Randomise analyses. Would we have to correct for this and adjust the thresholds accordingly?
I read in one of your previous posts on a related topic the following: “The strict answer is that yes, if the 4 tests were all independent then it would strictly be correct to add an extra factor of 4 to the correction. However, a) if they are not independent of each other then this is over-conservative and b) AFAIK no-one actually does do this…….”
Re: Reg Multiple comparison correction.......P value significance level.....
Based on the current literature regarding DTI, I would say that these parameters (FA, MD, AD, RD) are not fully independent of each other. Hence, adjustment of thresholds might be over-conservative. Do you agree?
Best,
Moji
|