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
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