 |
 |
Hi Yuwal, It seems you have 2 treatment groups, and in each group, each participant undergoes 2 levels (pre/post). It's a case for a 2x2 mixed design. It has appeared a number of times in the mailing list. Search for "design_ekaterina.ods" and the comments therein (the more recent ones will have the updated link). All the best, Anderson On Tue, 13 Nov 2018 at 05:29, Yuval Argaman <[log in to unmask]> wrote: > Dear FSL experts and community, > > I'm but a novice in FSL, and I have a dilemma as in how to define the VBM > and TBSS design matrix. > > My study design is randomised, placebo-controlled, crossover and > counterbalanced. > > Each of the 26 participants in my study underwent 4 scans: Pre-Treatment1, > Post-Treatment1, Pre-Treatment2 and Post-Treatment2. > > Initially, my go-to design was a 2x2 mixed effects ANOVA (2 treatments, 2 > timepoints). > > However, I'm not sure if this is the correct way to define the design > matrix, because the 1-factor X 4 measures might be more appropriate to use, > and seems simpler. > > Therefore, I'd appreciate your opinion regarding the following questions: > > 1. Assuming a >=4 weeks washout period between treatments, and a ~2-week > interval between scans in each treatment - which of the designs would be > more > appropriate? (I'm also performing a paired t-test to examine carryover > effects). > > 2. Is it necessary to define exchangeability blocks for each subject? > > Thanks in advance! > > Sincerely, > > Yuval. > > ######################################################################## > > To unsubscribe from the FSL list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 > ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1