Dear Jesús,
On Wed, 23 Aug 2017 12:26:43 +0100, Jesús Adrián-Ventura <[log in to unmask]> wrote:
>Hi everybody,
>
>I'm reporting the results of a multiple regression analysis after applying TFCE (Smith & Nichols, 2009) correction (p<0.05 FWE corrected).
>Seeing the table results I observe both the "cluster-level" and "combined peak-cluster-level" columns (which includes the TFCE section).
>
>I assume that the TFCE column shows the k-size of the cluster, although with decimals (i. e. 30.55), but I don't understand the meaning of the previous k in the cluster-level column, since it neither coindices with the SPM results (without TFCE). How should I correctly report the TFCE results?
>
The TFEC-column reports the TFCE value that results from combining voxel-height and cluster size and is the value that is used for the stat. analysis and thresholding. However, your reported value of 30.55 seems to be much too low and should be rather in the range of > 1000.
The k_E value is indeed the cluster size that gets somewhat unnecessary for the statistics, but is of some use to check whether this is a larger cluster or just a very tiny one. Thus, it is just for indicating the extent of the cluster in order to find it in your image.
Best,
Christian
>
>Thanks in advance.
>
>Jesús
|