----- Forwarded message from
[email protected] -----
Date: Thu, 25 Oct 2012
13:07:56 -0700
From: [email protected]
Reply-To: [email protected]
Subject: Comparison
Allometric Trajectories in MorphoJ
To:
[email protected]
----- Forwarded message from
Javier Santos
Date: Thu, 25 Oct
2012 19:39:04 +0200
From: Javier Santos
Reply-To: Javier Santos
Subject:
Comparison Allometric Trajectories in MorphoJ
To: Morphomet New Mailing
List
Hello Morphometricians,
I am studying ontogentic trajectories in three Lake Victoria cichlids in MorphoJ. I want to compare their allometric trajectories and have a doubt on how to proceed correctly. Currently the procedure I have followed is to do a regression of shape on lnCS for one ontogenetic stage using all three species. Then I exported the regression scores and did a bootstrap procedure to test differences between slopes across species. I think that this is the correct procedure, however I wanted to use the Compare Vectors function in MorphoJ. To do so, I divided the dataset for the same ontogenetic stage by species and did a seperate regression of shape on lnCS for each species. Afterwards I used the Compare Vector function to test differences between species' regressions. However, doing the comparison in this way implies that I had to do a seperate Procrustes fit for each species, which I think makes the comparison not statistically correct. Also, I wouldn't know how to interpret the angles obtained between species' vectors. So, my question is: Is there a way to use the Compare Vectors function to test for differences between species within the same regression analysis (shape on lnCS) in MorphoJ? and if not, why is it incorrect to proceed in that way? Thanks for your help.
Cheers,
Javier Santos Santos
[email protected]
Collaborator @ Department of Biodiversity and Evolutionary Biology
- Museo Nacional de Ciencias Naturales (MNCN-CSIC)
Associated Researcher @ Evolutionary Morphology of Vertebrates
- Universiteit Gent
I am studying ontogentic trajectories in three Lake Victoria cichlids in MorphoJ. I want to compare their allometric trajectories and have a doubt on how to proceed correctly. Currently the procedure I have followed is to do a regression of shape on lnCS for one ontogenetic stage using all three species. Then I exported the regression scores and did a bootstrap procedure to test differences between slopes across species. I think that this is the correct procedure, however I wanted to use the Compare Vectors function in MorphoJ. To do so, I divided the dataset for the same ontogenetic stage by species and did a seperate regression of shape on lnCS for each species. Afterwards I used the Compare Vector function to test differences between species' regressions. However, doing the comparison in this way implies that I had to do a seperate Procrustes fit for each species, which I think makes the comparison not statistically correct. Also, I wouldn't know how to interpret the angles obtained between species' vectors. So, my question is: Is there a way to use the Compare Vectors function to test for differences between species within the same regression analysis (shape on lnCS) in MorphoJ? and if not, why is it incorrect to proceed in that way? Thanks for your help.
Cheers,
Javier Santos Santos
[email protected]
Collaborator @ Department of Biodiversity and Evolutionary Biology
- Museo Nacional de Ciencias Naturales (MNCN-CSIC)
Associated Researcher @ Evolutionary Morphology of Vertebrates
- Universiteit Gent
----- End forwarded message -----
----- End forwarded message
-----
