-------- Original Message --------
Subject: Two questions regarding CVA
Date: Fri, 18 Feb 2011 11:30:19 -0500
From: Madeleine Chollet <[email protected]>
To: [email protected]
I am currently working on a project in which I am comparing brain shape
among several groups. I used MorphoJ to conduct a CVA to test for
statistical differences among the (3+) groups. I have two questions:
(1) The output includes p-values for both Procrustes distances and
Mahalanobis distances. What dictates when one should use Mahalanobis
distances versus Procrustes distances? I have non-significant results
for Procrustes distances but very significant results (i.e., p < 0.001)
for Mahalanobis distances.
(2) Is there a way to calculate how much of the total (sample) variance
is explained by differences identified in CVA?
Thank you for your help!
Madeleine
[email protected]
