-------- Original Message --------
Subject: Re: variance in shape
Date: Mon, 2 Jun 2008 08:11:03 -0700 (PDT)
From: Philipp Mitteroecker <[EMAIL PROTECTED]>
To: morphmet@morphometrics.org
References: <[EMAIL PROTECTED]>
Dear Chris,
CVA standardizes the within-group covariance structure to a spherical
distribution with unit variances along each axis. It would thus be
greatly misleading to compare variances of CVA scores.
Use the original Procrustes coordinates (or PCA scores or PW scores) to
compute variances. There are two common ways to produce a single summary
of variation in a multivariate data set: (1) the trace of the covariance
matrix (sometimes called total variance), which equals the sum of all
univariate variances or the sum of all eigenvalues, (2) the determinant
of the covariance matrix (generalized variance), which equals the
product of all eigenvalues. The latter one relates to the "volume" of
the point cloud, requiring a full rank covariance matrix. In an earlier
comment on morphmet Jim Rohlf gave more details on that.
Best,
Philipp
On Mo, 2.06.2008, 14:58, morphmet wrote:
-------- Original Message --------
Subject: variance in shape
Date: Sun, 1 Jun 2008 16:41:20 -0700 (PDT)
From: Chris Harrod <[EMAIL PROTECTED]>
To: morphmet@morphometrics.org <morphmet@morphometrics.org>
Dear all out there in morpholand.
I want to examine variance in shape between fish found in allopatry and
those found in sympatry with a putative competitor.
Can I simply compare the variance in DF/CVA scores (based on the normal
landmark analyses) between different populations, or am I being horribly
simplistic?
thanks in advance
Chris
<> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>
Dr Chris Harrod
Lecturer in Fish and Aquatic Ecology
Ecology and Evolutionary Biology
Queen's University Belfast
School of Biological Sciences
Medical Biology Centre
97 Lisburn Road
Belfast BT9 7BL
UK
[EMAIL PROTECTED]
Tel: +44 (0) 28 90972271
Mob: +44 (0) 79 77419314
Fax: +44 (0) 28 90975877
http://www.qub.ac.uk/schools/SchoolofBiologicalSciences/Staff/DrCHarrod/
<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
--
Dr. Philipp Mitteröcker
Konrad Lorenz Institute for Evolution and Cognition Research, Austria
Department of Anthropology, University of Vienna, Austria
http://www.virtual-anthropology.com/Members/philippm
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
