Hi NBS community

I would like to verify whether my design matrix for one group with four different conditions (repeated measures) has been set up correctly. I will be happy about a quick comment on it and thank you in advance for any kind of advice.

Information about the design:
13 subjects, 4 tasks A, B, C, D
Research question: task A vs tasks B, C, D

Design matrix:
I have attached the design matrix. The connectivity matrices/data files of the four tasks are ordered one below the other (subjects 1-13 for task A, subjects 1-13 for task B, subjects 1-13 for task C, subjects 1-13 for task C). The first four columns identify the four tasks, the following columns indicate repeated measures. Is this correct? Is such a design admissible to NBS?

The contrast according to my research question would be (Task A vs Task B, C, D): 1 -0.33 -0.33 -0.33 0 0 0 0 0 0 0 0 0 0 0 0 0
I set up the exchange blocks for repeated measures: 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13

Statistical test to run the analysis: ANOVA

I'm looking forward to your answers.
Kind regards,
Julia

Hi Julia, 

The way you have set up your exchange blocks is ok. However, your design matrix is rank deficient and won't work. I suggest that you take at look at the tripled t-test design matrix example described here:

http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FEAT/UserGuide#Tripled_Two-Group_Difference_.28.22Tripled.22_T-Test.29

This provides an example for the case of three tasks. I suggest following this design matrix.You will need to add another column because that example is for three tasks, but you have four.

Alternatively, you can start simple, and test each of the three pairs (A vs B, A vs C, A vs D) independently using a paired t-test.

Andrew


Originally posted by Julia Binder: