Dear list-members --
I am relatively new to this list (after enjoying the archives for a while, I
decided to add myself in) -- this is my first foray into actually posting,
and am a bit on the intimidated side, so please forgive my question if it is
elementary...
I am currently trying to interpret GROUP differences in molar occlusal shape
using 2-D TPS (I have been using the TPSdig and IMP packages) and have been
using the partial warp scores as variables in CVA and DFA analyses. In
particular, I am trying to see if the partial warp scores can be used in my
particular sample to discriminate between groups, and if they can be used as
predictor variables (for interpreting diet in fossil taxa based on shape in
modern taxa).
My question is as follows:
I have been searching all of the "usual" bibliographic databases for papers
that specifically use CVA (over PCA) and attempt to use partial warp scores
in actual DFA classifications. I have been unable to really come up with
anything, and was wondering if anyone could point me in the correct
direction. I assume that I am probably missing it along the way! I realize
that many people use PCA (as they are interested in individual differences),
but I am specifically interested in group mean differences in shape. Am I
wrong in assuming that partial warps can even be used in such a manner?
As a follow-up: can partial warp scores derived from 3-D data also be used
in the same way (i.e., in CVA and DFA)?
Thank you so much in advance for any help that might be sent my way. Again,
I apologize if my question is too elementary on this subject...
Jess White
Visiting Professor
Department of Sociology and Anthropology
Western Illinois University
Macomb, IL 61455
[EMAIL PROTECTED]
[EMAIL PROTECTED]
_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org
