External Email - Use Caution        

Dear experts,

 

I am conducting a group analysis and wanted to confirm that the set up below accurately addresses our research question. Thanks for your help, much appreciated!

 

Background:

• We have two groups (intervention vs control) and two timepoints (pre vs post)
• Our data have already been processed and stacked with long_mris_slopes where pc1 is the measure of interest (post - pre).
• Our covariates are sex (male vs female), age, and MMSE scores.
• Thus, our model has 4 groups (2 factors/2 levels):
  • int_male
  • int_female
  • con_male
  • con_female
• We’re using DODS
• We’re running the analysis with mri_glmfit in the Longitudinal Two Stage Model

 

As such, we want test whether there are greater changes in the intervention group vs control group in the pc1 measure, adjusting for the influence of sex, age, and MMSE.

 

This is our FSGD:

GroupDescriptorFile 1

Title Between-group

Class int_male

Class int_female

Class con_male

Class con_female

Variables age mmse

Input 001.MR int_male      30     25

Input 002.MR int_female    40     22

Input 003.MR con_male      60     29

Input 004.MR con_female    60     21

                …

 

And this is the contrast matrix

+0.5 +0.5 -0.5 -0.5 +0 +0 +0 +0 +0 +0 +0 +0

 

With the above contrast matrix, our interpretation would be:

• Intervention > control = positive result
• Intervention < control = negative result

 

Is this the correct set up and interpretation?

 

Thanks for taking the time,

Best regards,

Nárlon Cássio

 

Nárlon Cássio Boa Sorte Silva PhD (He, Him, His)

CIHR and MSHR BC/PARF Postdoctoral Research Fellow

Aging, Mobility, and Cognitive Health Lab

Djavad Mowafaghian Centre for Brain Health

Department of Physical Therapy, Faculty of Medicine

University of British Columbia

Vancouver, Canada

Twitter: @BoaNarlon

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer