help > RE: Longitudinal Design
May 27, 2020  04:05 PM | Andrew Zalesky
RE: Longitudinal Design
Hi Giulia,

Please note that the age of your subjects does not change between the two points. Moreover, the gender also does not change between the two points. Therefore, the columns corresponding to age and gender should be removed. If you leave these columns in the matrix it will cause a rank deficient error because the columns are not linearly independent.

There is no need to control for confounds that do not change between the two time points. These confounds will be accounted for in the within-subject means.

If you remove these two columns, I think that the design should work ok.

Andrew


Originally posted by Giulia Tronchin:
Dear Prof. Zalesky,

I have a longitudinal analysis to run, and I would like to check the group*time interaction (2 groups, 2 time points). This is how I designed my design matrix and contrast. Unfortunately, I am getting the Rank deficient error in matlab and I am not sure why.


0 0 1 52 1 0 0 0 0 0
0 0 0 48 0 1 0 0 0 0
0 0 1 41 0 0 1 0 0 0
0 0 1 45 0 0 0 1 0 0
0 0 1 48 0 0 0 0 1 0
0 0 0 44 0 0 0 0 0 1
1 0 1 52 1 0 0 0 0 0
1 0 0 48 0 1 0 0 0 0
1 1 1 41 0 0 1 0 0 0
1 0 1 45 0 0 0 1 0 0
1 0 1 48 0 0 0 0 1 0
1 0 0 44 0 0 0 0 0 1

Contrast: 0, 1, 0, 0, 0, 0, 0, 0, 0, 0
F- test

1st column is time, 2nd column is group*time interaction, 3rd column gender and 4th column age and the other columns are for each of 6 subjects.

Thanks for your help with this,
Giulia

Threaded View

TitleAuthorDate
Giulia Tronchin May 27, 2020
RE: Longitudinal Design
Andrew Zalesky May 27, 2020
Giulia Tronchin Jun 2, 2020