Re: [R] centroid representation and MANOVA

y Hamilton, Ontario, Canada http://socserv.mcmaster.ca/jfox > -Original Message- > From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] > On Behalf Of julien colomb > Sent: April-27-11 10:41 AM > To: r-help@r-project.org > Subject: [R] centroid repre

[R] centroid representation and MANOVA

hi all. I have a matrix of data with 5 different groups and 20 individual response per group, and about 12 variables collected for each. I want to represent the result in a 2D plot. PCA is not so good because the difference between the groups is not obvious. I have seen, in a recent paper, pe