Correlation of DVs in MANOVA

From: NWM (pattgeary_at_hotmail.com)
Date: 11/25/04 

Date: 25 Nov 2004 11:58:03 -0800

Hi,

Two quotes from "Using Multivariate Statistics" by Tabachnick 4th ed.

"The best choice is a set of DVs that are uncorrelated with each other
because they each measure a separate aspect of the influence of the
IVs ... What is gained by inclusion of several measures of the same
thing?" P328

And,

"MANOVA works best with highly negatively correlated DVs, and
acceptably well with moderately correlated DVs in either direction ...
MANOVA is less attractive if correlations among DVs are very highly
positive or near zero ... MANOVA is also wasteful if DVs are
uncorrelated". P357

So, which is correct and how to reconcile her two statements?

Thanks!

Quantcast