Hi Torben,

Thank you for your answer. Following it, I browsed the Archives to see how to check the residuals. In fact there is an answer very recently posted by Tom Nichols regarding the extraction of the residuals (http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0804&L=SPM&P=R60255&I=-3). I would like to confirm if this is the procedure I have to apply to check the residuals' distribution. I work on a VBM study (I use spm2).

Thank you again,

Rachel

--- On Sun, 4/20/08, Torben Ellegaard Lund <[log in to unmask]> wrote:

From: Torben Ellegaard Lund <[log in to unmask]>
Subject: [SPM]
To: [log in to unmask]
Date: Sunday, April 20, 2008, 3:39 AM

Hi Rachel

It is the residuals which need to be independent and identically (normal) distributed, not the covariates of interst. 

Hope this helps

Torben

Torben Ellegaard Lund

Assistant Professor, PhD

The Danish National Research Foundation's Center for Functionally Integrative Neuroscience (CFIN)

Aarhus University

Aarhus University Hospital

Building 30

Noerrebrogade

8000 Aarhus C

Denmark

Phone: +4589494380

Fax: +4589494400

http://www.cfin.au.dk

[log in to unmask]

Den 20/04/2008 kl. 05.32 skrev Rachel Grossman:

Hi SPM community,   I have a question concerning the distribution of covariates of interest. How does SPM deal with covariates of interest that are not normally distributed? For instance if the values are skewed toward the minimum and the maximum limits?   Thanks for your advice, Rachel

---

Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.