It sounds like you've done the right thing. It is not too unusual for all/most of your points to disappear when you correct for multiple comparisons. You might try a monte carlo simulation instead (do a search on the FreeSurfer wiki to find the page).
doug
jitesh kar wrote:
Dear Free surfer Mailing List, This is my first mail as recently I have started using free surfer. I am a research assistant in Neurology, at University of Alabama at Birmingham. I do not have much knowledge about free surfer(I will really appreciate if your reply is simple and has less technical terms) but I try to go through tutorials or free surfer wiki if I have some query. We have some data on picture response time(as a part of sentence comprehension) and we want to correlate the effect of aging on this response time. We are comparing gray matter thickness with picture response time. I tried to run qdec but I am still confused with how to use FDR function? Suppose I select a variable (e.g Picture response time) and run correlation with gray matter thickness, it gives me 4 options:
- Does thickness differ from zero?
- Does picture response time correlates with thickness?
- standard deviation
- correlation coefficient
Now, if I click on no.2 I do get some points on brain which correlates with my variable, does that mean it is significant findings? or do I have to click on "set using FDR"? The problem with using FDR is all my points disappears. I am sorry but I do not have much technical knowledge about free surfer or statistics but I am curious to learn about this. Is it mandatory to click on FDR?
- JK
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer