help > Within-subject analysis for 1 group
Showing 1-2 of 2 posts
Nov 6, 2017 11:11 AM | Yi-Chia Kung
Within-subject analysis for 1 group
Dear Dr. Zalesky,
I would like to do the one way ANOVA within-subject analysis for N subjects with three time-points in one group.
I have tried some design matrix and the corresponding contrast in the following: (2 subjects and 3 time-points for example)
Design matrix:
[1 0 1 -1 0;
0 1 1 -1 0;
1 0 0 1 -1;
0 1 0 1 -1;
1 0 -1 0 1;
0 1 -1 0 1]
Contrast:
[0 0 1 1 1]
Statistical Test: One-sample
However, there is a "Rank deficit" warning.
Is it be possible to do the one-way ANOVA within-subjects (and the following post-hoc) analysis with NBS?
Thank you very much!
Best,
Clementine
I would like to do the one way ANOVA within-subject analysis for N subjects with three time-points in one group.
I have tried some design matrix and the corresponding contrast in the following: (2 subjects and 3 time-points for example)
Design matrix:
[1 0 1 -1 0;
0 1 1 -1 0;
1 0 0 1 -1;
0 1 0 1 -1;
1 0 -1 0 1;
0 1 -1 0 1]
Contrast:
[0 0 1 1 1]
Statistical Test: One-sample
However, there is a "Rank deficit" warning.
Is it be possible to do the one-way ANOVA within-subjects (and the following post-hoc) analysis with NBS?
Thank you very much!
Best,
Clementine
Nov 7, 2017 12:11 AM | Andrew Zalesky
Within-subject analysis for 1 group
Hi Clementine,
this design matrix is not correct and this is not a one-sample text. For a 1 x 3 repeated measures ANOVA, use the following design matrix
1 0 1 0 %subject 1 time 1
0 1 1 0 %subject 2 time 1
1 0 0 1 %subject 1 time 2
0 1 0 1 %s2 t2
1 0 0 0 %s1 t3
0 1 0 0 %s2 t3
contrast: [ 0 0 1 1]
exchange blocks [ 1 2 1 2 1 2]
test: ftest
A more detailed example can be found here:
https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/F...
Andrew
Originally posted by Yi-Chia Kung:
this design matrix is not correct and this is not a one-sample text. For a 1 x 3 repeated measures ANOVA, use the following design matrix
1 0 1 0 %subject 1 time 1
0 1 1 0 %subject 2 time 1
1 0 0 1 %subject 1 time 2
0 1 0 1 %s2 t2
1 0 0 0 %s1 t3
0 1 0 0 %s2 t3
contrast: [ 0 0 1 1]
exchange blocks [ 1 2 1 2 1 2]
test: ftest
A more detailed example can be found here:
https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/F...
Andrew
Originally posted by Yi-Chia Kung:
Dear Dr. Zalesky,
I would like to do the one way ANOVA within-subject analysis for N subjects with three time-points in one group.
I have tried some design matrix and the corresponding contrast in the following: (2 subjects and 3 time-points for example)
Design matrix:
[1 0 1 -1 0;
0 1 1 -1 0;
1 0 0 1 -1;
0 1 0 1 -1;
1 0 -1 0 1;
0 1 -1 0 1]
Contrast:
[0 0 1 1 1]
Statistical Test: One-sample
However, there is a "Rank deficit" warning.
Is it be possible to do the one-way ANOVA within-subjects (and the following post-hoc) analysis with NBS?
Thank you very much!
Best,
Clementine
I would like to do the one way ANOVA within-subject analysis for N subjects with three time-points in one group.
I have tried some design matrix and the corresponding contrast in the following: (2 subjects and 3 time-points for example)
Design matrix:
[1 0 1 -1 0;
0 1 1 -1 0;
1 0 0 1 -1;
0 1 0 1 -1;
1 0 -1 0 1;
0 1 -1 0 1]
Contrast:
[0 0 1 1 1]
Statistical Test: One-sample
However, there is a "Rank deficit" warning.
Is it be possible to do the one-way ANOVA within-subjects (and the following post-hoc) analysis with NBS?
Thank you very much!
Best,
Clementine