AFNI Message Board

Dear AFNI users-

We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:

https://discuss.afni.nimh.nih.gov

Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.

The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.

Sincerely, AFNI HQ

History of AFNI updates  

Previous Message Next Message
Message List New Topic Print View
|
gang
April 18, 2013 10:59AM
Hi Andy,

My personal opinion is that the assumption of homogeneity of regression for ANCOVA is really an old school stuff. Without knowing (or testing) whether the groups have the same covariate effect, a prior assumption seems a little foolhardy.

Having said that, 3dttest++ and 3dMEMA both have the capability to model the same or different covariate effect across groups. 3dMEMA has the option -covariates_model slope=different/same for that purpose.

Alternatively, you can pretend that you have only *one* group, and then dummy code your groups as a covariate. That way you can decide you want to model as well as test the groups with the same or different covariate effect in 3dttest++ or 3dMEMA. More specifically if you want different covariate effect across groups, add an interaction column in the covariate file. With 0/1 coding, you can check whether the groups have significant different covariate effect.

Gang



Edited 1 time(s). Last edit at 04/18/2013 12:31PM by Gang.
Reply Quote
RSS
Subject Author Posted

3dttest++ and homogeneity of regression

Anonymous User April 17, 2013 07:24PM

Re: 3dttest++ and homogeneity of regression

gang April 18, 2013 10:59AM