Hi Matt,

For n subjects, the number of possible permutations in the paired t-test is 2^n. It's the same as you'd have if you subtracted the differences and did as a 1-sample t-test on these differences.

You'd use one exchangeability block per subject.

All the best,

Anderson


On Wed, 17 Apr 2019 at 11:50, Sherwood, Matt <[log in to unmask]> wrote:

FSL community,

 

I am running a paired t-test using randomise on data in which I have 2 samples from a single group of individuals. I know the proper way to conduct this would be to calculate the delta between the samples and send that as a 1 sample test. However, I am interested in running this and a traditional paired t-test where the exchangeability blocks are defined appropriately. In the case of a 1 sample test, the number of permutations possible is 2^n. Is this also the case for the paired t-test or do I follow the equation for the two-sample test?

 

Thanks

_Matt

 

Matthew Sherwood, Ph.D.

Research Engineer II

Department of Biomedical, Industrial & Human Factors Engineering

Wright State University | 3640 Colonel Glenn Hwy. | Dayton, OH 45435

Located in 413 Neuroscience Engineering Collaboration

 

cell: (937) 503-7178 | fax: (937) 775-7364

[log in to unmask] | wright.edu/neumap

 


To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1


To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1