 |
 |
Hi Kim, Although it is possible in theory, in practice it will be much more complicated, particularly when computing the varcopes, that need the design and the contrast that was omitted. It is probably less work and definitely less error prone to simply run the analysis again, even if it may take a couple of days. All the best, Anderson On 15 May 2017 at 09:54, Kim Li <[log in to unmask]> wrote: > Hi All, > > I have completed the individual level analysis and realized that I made a > mistake in specifying one of the contrast. I have 3 EVs: female, male and > object; for contrasts, I did female, male, object, female-object. However > I also need male-object contrast now. What is the best way to acquire > male-object contrast without running the entire analysis again? I figured > since I have male and object copes on individual level, I should be able to > obtain male-object somehow? > > I was thinking maybe I could do a intermediate level to combine the two > copes? > In FEAT GUI, use higher-level analysis, input 3D copes, and use fixed > effect models to make (using 4 subjects as example, top is male copes and > bottom -1 are for object copes): > 1 0 0 0 > 0 1 0 0 > 0 0 1 0 > 0 0 0 1 > -1 0 0 0 > 0 -1 0 0 > 0 0 -1 0 > 0 0 0 -1 > and make contrast as > 1 0 0 0 > 0 1 0 0 > 0 0 1 0 > 0 0 0 1 > so that each contrast would create a cope for each subject that is > male-object. > > I wonder if that is a possible way to combine the values. If not, could > anyone help me to figure out how to do that? Thanks ahead of time. > > Sincerely, > > Kim >