open-discussion > Meaning of T statistic value in second level analysis
Showing 1-2 of 2 posts
Display:
Results per page:
Jul 5, 2019  08:07 AM | ate - IIS La Fe
Meaning of T statistic value in second level analysis
Dear all,

I would like to ask about the T statistic mean in the second analysis. After launch a hypothesis and obtain the regions with significant difference, we obtain different statistics below the Analysis results.

Please, could you explain me o refer me to some paper / manual where the mean of T value and the number into the parenthesis are explained.
I suspect that T statistic value means the difference between two population compared (e.j Patients vs Healthy subjects) in standard deviation units but I would like to confirm o reject this opinion.
Kind regards,
Jul 8, 2019  06:07 AM | Steve Hodge
RE: Meaning of T statistic value in second level analysis
Hello,

Your suspicion is probably correct: the t-statistic is something like the difference between two sample means, scaled by their standard error (and there are several ways to estimate the standard error).

But without knowing the software program and the t-statistic you are referring to ("the mean of T value and the number into the parenthesis"), I cannot be sure it is the correct interpretation for your results. It could be that you are seeing the average t-statistic for a region of interest (and the associated measure of variability among the t values). Examining the basic descriptive statistics between your groups may be helpful.

Some more information can be found here:
T-statistic in general:
https://en.wikipedia.org/wiki/Student%27...

T-statistic as contrast in neuroimaging:
http://mindhive.mit.edu/node/60f

Steve
---
Steve Hodge