Strictly speaking, because multiple comparisons were involved here, correction is necessary. On Thu, Mar 3, 2016 at 9:18 AM, SUBSCRIBE FSL Alain Ridet < [log in to unmask]> wrote: > Hi, > I have a question that I have a lot of difficulties concerning statistics. > Indeed, I extracted the FA values between 2 experimental conditions. I have > paired comparisons between pretest-posttest with my 2 Regions of interest > to the two hemispheres. I would like to know if in my case I have to > correct my p-value. Do I "necessarily" have to divide my p value by 4 > although these are areas that are part of my hypotheses a priori? > Thank you in advance for any help and your time > -- ----------------------------------------------------------------------------------------- Kong Xiangzhen State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China, 100875.