Hi Matthew,
Thanks Matthew, Yes. Numbers of *glm_pe_tstat* is based on the contrast.
I wonder if I get your propmt:
Here is my design matrix, contrast and design,fsf in the attachment.
In my case, I completed the correlation. And I found the second contrast(i.e. score_year) has a corrected significant result, so I want to know whether the the correlation is negtive or positive.
I think I can get the beta value in the 2nd volume of each file. If so, the value of 2nd volume in each file should be the same. But it does not in my files. That is why? I also want to know which beta values represent 5 volumes in one file(i.e *_glm_pe_tstat1.nii.gz) respectively .
Thanks.
All the best.
Rujing Zha
At 2014-01-03 21:27:03,"Matthew Webster" <[log in to unmask]> wrote:
Hello Rujing,
The number of t-stat files produced will be equal to the number of contrasts. Each contrast will output the GLM information in the appropriately numbered file. These files will contain the PE values in numerical order - in this case the 2nd image in each file will be the PE for the second regressor.
Kind Regards
Matthew
Dear Anderson and others,
I got the result of glm output by --glm_output option in randomise. And I want to know whether regressor positively or negtively correlated GM data in significant corrected areas.
As I had 5 regressors, I got five *_glm_pe_tstat* files. Because of the 2th regressor had a corrected regions, I viewed *_glm_pe_tstat2.nii.gz. However this file had contained 5 volumes, I wonder from which volume I could get the beta value to know it was positive or negtive.
Thanks Andrson and others.
All the best.
Rujing Zha
