----- Forwarded message from Robert Richardson -----
Date: Fri, 1 Jun 2012 17:42:44 -0400
From: Robert Richardson
Reply-To: Robert Richardson
Subject: multi-class ANOVA analogue for geometric morphometrics
To: morphmet-requ...@morphometrics.org
Fellow morphers,
I'm familiar with the need to account for the lower degrees of
freedom, when testing for differences between groups, hence the use of
Goodall's-F test. However, as with basic (M)ANOVA, that just tells
you if there's at least one group that's statistically different but
doesn't identify which one(s) when you have more than two categories.
Are there multi-class (M)ANOVAs (e.g. Ryan-Einot-Gabrial-Welsch Q-test
or Tukey's test) that are out there that will directly identify which
categories are different when dealing with many categories (e.g.
samples from seven geographical locations)? Maybe there's a way to
load the data into an R-script and manually adjust the degrees of
freedom for a regular multi-category (M)ANOVA?
Thanks for the input!
Robert-
--
=== === === ===
Robert Richardson, PhD candidate
Department of Biology
Portland State University
1719 SW 10th Ave; SB 2, #246
Portland, OR 97201
503.725.8004; r...@pdx.edu
=== === === ===
----- End forwarded message -----
