pair t test, usually there is only one sample per subject

[zqin1998]

When I use matlab palm, I want to do a pair t-test, but I don't average the n samples of a subject, but input them directly, so that my design matrix will have repeated 1's under a subject. There will be no bugs. What analysis should I run at this time? design matrix like this 1 | 1 | 0 1 | 1 | 0 1 | 1 | 0 1 | 1 | 0 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 1 | 0 | 1 -1 | 1 | 0 -1 | 1 | 0 -1 | 1 | 0 -1 | 1 | 0 -1 | 1 | 0 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1 -1 | 0 | 1

contrast matrix like this

1 | 0 | 0 -1 | 0 | 0

pair t test, usually there is only one sample per subject

[andersonwinkler]

Hi, I'm not sure I follow the question but you can start a design for a paired t-test using as example those from the FSL website: https://fsl.fmrib.ox.ac.uk/fsl/docs/#/statistics/glm?id=single-group-paired-difference-paired-t-test All the best, Anderson