Dear Oliver,
if I understood correctly your question, I have a feeling you cannot
do that in MorphoJ. There are probably ways to do it in R/Mablab.
However, again if I understood correctly your question, what you want
to test is perhaps best accomplished with a linear model (and similar
approaches) including a term for the interaction between treatment and
fish group.
Best,
Carmelo
[email protected] ha scritto:
I am currently trying to test the degree of difference (measured as
a Mahalanobis distance) between two groups.
To give some background I have 2 groups of fish that differ in their
life history strategy (groups A and B), these two groups have then
been exposed to different conditions to induce a phenotypic response
(conditions 1 and 2). I have then compared these groups in MorphoJ
using a single canonical variate analysis.
The out put gives me a Mahalanobis distance for all comparisons but
the ones I am interested in are for fish group A1 and A2 and for the
fish group B1 and B2.
The question I want to answer is one group more plastic in their
response to the 2 different conditions - So I need to test if the
Mahalanobis distance between A1 and A2 fish is significantly more,
less or not significant than the Mahalanobis distance between B1 and
B2, fish but there is no variation around the Mahalanobis distance
to allow a test..
Is there a way of extracting all the Mahlanobis distances for each
individual from MorphoJ so I can then test these in a linear model
or equivalent?
Or does anyone know a way of doing this?
Thanks in advance
Oliver Hooker
PR~Statistics
--
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
--
Carmelo Fruciano
Marie Curie Fellow - University of Konstanz - Konstanz, Germany
Honorary Fellow - University of Catania - Catania, Italy
e-mail [email protected]
http://www.fruciano.it/research/
--
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
