[MORPHMET] the problem with CVA... or is it?

Dear Morphmet-ers, I'm seeking advice on methods for visualizing shape features that distinguish multiple groups using GM. I know CVA has fallen out of favor for a number of reasons discussed here - e.g., more variables than groups, nonisotropic variation: Mitteroecker, P., and Bookstein, F.

Re: [MORPHMET] the problem with CVA... or is it?

Dear Christy, I'm not sure I understood the part on doing it "using specimens and not species". As you know, Mitteroecker & Bookstein (2011) suggest between-group principal components. Personally, I have used this technique multiple times (e.g., Fruciano et al 2014 - Biological Journal of

Re: [MORPHMET] Re: the problem with CVA... or is it?

Christy Hipsley ha scritto: Sorry I should have been more clear - the CVA was done using individual shapes, so n=161, and the bgPCA was on species means (the basic unit of my study), so n=92. I did the CVA on the individuals so as not to have more "groups" than

RE: [MORPHMET] the problem with CVA... or is it?

While you do have more sample than variables, it is too close. One prefers to have many more samples than the number of variables. Another problem is that a CVA with 144 variables will mostly involve a covariance matrix that is close to being singular. One way I like to check for problems (and