[The original submission had embedded graphics that do not show up in
the text-only version below. -the morphmet mod (dslice) ]
-------- Original Message --------
Subject: help
Date: Thu, 14 Jul 2011 18:19:59 -0400
From: Julien Sastre <[email protected]>
To: [email protected]
Hi,
I'm a new to geometric morphometrics analysis and I'm working on cranial
growth.
In order to assess the reliability of the method, landmarks have been
positionned twice by the same observer and twice by two differents
observers.
I would like to perform a Procrustes ANOVA in order to know if there is
differences between the two measurement.
For the reproductibility, I made a classifier "individual" where the
same individuals are presents twice and a classifier "measure_serie"
numbered 1 (=first measurement) or 2 (=second measurement).
Then I perfomed a Procruste Anova
And I obtained
Procrustes ANOVA: Procrustes ANOVA ...
Dataset: reproductibilité
Classifiers used for the Procrustes ANOVA:
Individuals: individual
Error 1: measure_serie
Centroid size:
EffectSSMSdf F P (param.)
Individual 2649,306645 294,367405 9 450,34 <.0001
Error 1 6,536558 0,653656 10
Shape, Procrustes ANOVA:
EffectSSMSdf F P (param.) Pillai tr. P (param.)
Individual 0,07606574 0,0000676140 1125 0,02 1,0000
Error 1 4,05229641 0,0032418371 1250
And I conclued that the two measurements are not significantly differents.
Is the method valid?
Any help is great.
Thanks.
Julien
