-------- Original Message --------
Subject: MANOVA with 2 views for each specimen ?
Date: Mon, 29 Dec 2008 13:14:54 -0800 (PST)
From: jay carroll <[email protected]>
To: [email protected]
Hello all. I have a data set which comprise two views (dorsal and a
lateral) for each individual. Both views have been digitized and
partial warps&uniform components have been generated. I have performed
a partial least square analysis to investigate the covariation between
the two views. But my question is focused on the differences between
different ecological groups. My first inclination would be to perform a
MANOVA model. But would it be more appropriate to analyze this data
with two separate MANOVA models (dorsal and lateral)? Alternatively,
could both the dorsal and lateral partial warps and uniform components
be distilled down to a few principle components and analyzed in a single
MANOVA model? Is this approach off base? Any special considerations
since both dorsal and lateral views are composed of two separate shape
spaces? Any comments on this idea or relevant citations would be
appreciated.
Jay ([email protected] <mailto:[email protected]>*)*
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
