Hi,
I am an undergraduate student and a new SPM user and have a few basic questions regarding analysis of results in SPM.
1) Shoudl I use t scores or Z scores in my results ? I read soemwhere that Z scores are derived form t scores by some conversion. However the Z scores give me ' inf ' for a few of my results. On the other hand , I get values for t scores for all my results which seems a better option.
2) Regarding the use of t scores for analysis, I get worried explaining stats since i am weak in it. I gather(I may be wrong) that a t score which is higher for one cluster compared to the other doesnt mean that the voxels in one cluster are more activated than the other cluster but it is to do with statistical signficance so I was wondering what terminology I should use when explaining my results.
3) For group analysis I have used not used FWE correction. Is it still ok to use the stats in this case for comparisons even though no correction is used ?
4) Are the cordinates the centre of the cluster or the maximum t score voxel coordinate. Given that it is the centre is it possible to get the corrdinate of voxel with the higest t score or vice versa ? Which one is more commonly used ?
Thanks a lot
|