Print

Print


You won’t be able to add the covariates directly to a paired T-test design, because each subject has consistent age/gender across their two time points, so there is a collinearity issue between (a linear combination of) your subject effect columns in the design matrix and your covariate columns. However, what you can do it subtract the two time point images using imcalc and then perform a one-sample T-test on the differences with age/gender covariates.  A paired T-test is statistically just a one-sample T-test on paired differences.

 

--

  Mike Angstadt

  Research Computer Specialist / PANLab Lab Manager

  Department of Psychiatry / University of Michigan

  (734) 936-8229

 

 

From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of André Schmidt
Sent: Friday, July 08, 2016 7:52 AM
To: [log in to unmask]
Subject: [SPM] paired t test with age and gender as covariate

 

Dear experts,

 

I have a fMRI dataset from 18 subjects with two timepoints, placebo and treatment (within subject), with an interval of about two weeks. I would like to add age (some subjects are twenty, while others are around fifty and sixty) and gender as covariate in the paired t test design. Can you please help me how I can do that?

 

Many thanks

Andre

**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues