 |
 |
Hello Nick, All of the below looks to me, although it’s worth noting that the reason demeaning isn’t required for any of your contrasts is that the group mean is modelled by the first EV. Adding the -D option to randomise would be counterproductive here. Kind Regards Matthew -------------------------------- Dr Matthew Webster FMRIB Centre John Radcliffe Hospital University of Oxford On 2 Sep 2018, at 08:33, Niklas Lenfeldt <[log in to unmask]<mailto:[log in to unmask]>> wrote: pushing this question best nick I just want to check. I have a model where a want to check the correlation between a continous parameter A and FA correcting for sex and age. Only one group. The design file is then: group gender age A And group as all ones and then the cont. values for each patient (not demeaned as randomize does that for you). I am intereseted in the effect of A accountong for the effect of sex and age, giving a contrast file as 0 0 0 -1 (neg correlation assumed) Is this correct? Just to check, if I want the separate effect of sex and age corrected for the others, would that be: 0 0 0 -1 # A (corresponding to stat-files number 1) 0 1 0 0 # sex (coresp to no 2) 0 0 -1 0 # age (corresp to 3) Is this the way to interpret it, i.e. the stat-file numbering refers to the lines in the contrast file? best nick ________________________________ 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 ######################################################################## To unsubscribe from the FSL list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1