-------- Original Message --------
Subject: RE: multi-class ANOVA analogue for geometric morphometrics
Date: Mon, 4 Jun 2012 14:59:23 -0400
From: F. James Rohlf <ro...@life.bio.sunysb.edu>
Reply-To: ro...@life.bio.sunysb.edu
Organization: Stony Brook University
To: morphmet@morphometrics.org
Once you have projected your shapes into the tangent space (e.g.,
partial warp scores) then you have what can be treated as ordinary
multivariate data. You can use any multivariate method that is
appropriate for the questions you wish to ask.
Note that you should not normally use Goodall's test as it makes many
unrealistic assumptions and thus in practice has much higher Type I
error rates than you would expect.
----------------------
F. James Rohlf, John S. Toll Professor, Stony Brook University
The much revised 4th editions of Biometry and Statistical Tables are now
available:
http://www.whfreeman.com/Catalog/product/biometry-fourthedition-sokal
http://www.whfreeman.com/Catalog/product/statisticaltables-fourthedition-rohlf
Please consider the environment before printing this email
-----Original Message-----
From: morphmet_modera...@morphometrics.org
[mailto:morphmet_modera...@morphometrics.org]
Sent: Monday, June 04, 2012 2:05 AM
To: morphmet@morphometrics.org
Subject: multi-class ANOVA analogue for geometric morphometrics
----- 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 -----