Great, thank you!
> On Nov 14, 2013, at 4:25 PM, Mark Jenkinson <[log in to unmask]> wrote:
>
> Hi,
>
> I would advise you not to quote t-statistical values when using TFCE, as they are not standard t-statistics and could be misinterpreted. It is really the p-values that are the most meaningful, so I would say that you should stick with them.
>
> All the best,
> Mark
>
>
>> On 13 Nov 2013, at 23:33, Dana Wagshal <[log in to unmask]> wrote:
>>
>> I completed a VBM analysis following the userguide and then used randomise with 10,000 permutations using the threshold-free cluster enhancement option. I got significant results in my vbm_s3_tfce_corrp_tstat1.nii.gz image with the clusters thresholded at p < 0.05.
>>
>> For publication, I want to create a color bar that has the minimum and maximum t-statistic that produced significant results. I got the max by using fslstats but I'm unsure what to put as the minimum. Sorry if this is a basic question!!! And it is recommended that to put it terms of the t-statistic or as a z-value?
|