-------- Original Message -------- Subject: Re: Fluctuating asymmetry analysis Date: Thu, 26 May 2011 06:21:57 -0400 From: Chris Klingenberg <[email protected]> Reply-To: [email protected] Organization: University of Manchester To: [email protected] Dear Marion I suspect the problem might be with the classifier for "individual". You need a classifier that indicates the individual, with values such as "vole 1", "vole 2", "vole 3".... It is not the format that counts (it could be things like "Joe" or "v569_f 3"), but the classifier must have exactly the same value for each replicate (image) of each individual vole and the value for each particular vole must be different from that of all other voles. If there is no such classifier, you have to male one (e.g. extract it from the identifier, make one manually by edit Classifiers, or import it from a spreadsheet). That your classifier for the individual effect is called "from dataset" makes me suspect there may be a problem. MorphoJ creates a classifier of this name if you combine different datasets. This may be correct in your case if you had the landmark configurations from the replicate images of each vole in a separate dataset and then combined them all. If not, that's most likely where the problem is. I hope this helps. Best wishes, Chris On 5/26/2011 10:24 AM, morphmet wrote:
-------- Original Message -------- Subject: Fluctuating asymmetry analysis Date: Wed, 25 May 2011 11:46:24 -0400 From: marion chevrinais <[email protected]> To: [email protected] Hi, I'm a new to geometric morphometrics analysis and I try to work on the fluctuating asymmetry of vole's populations. For this, I make landmarks and I would use MorphoJ to realise a procrustes ANOVA. I have some voles to the North of Quebec and some to the south I would compare the fluctuating asymmetry of each population to characterize the stress which they are subject. In MorphoJ, I have made the procrustes fit, and after I have created a covariance matrice. I have also created some classifiers. One for the geographic location I have called "geography", I discriminated north to south population. And I created a classifier for the error of measurements I have called "image", I put 1 or 2 respectively for the first image and the replicate. When I would execute the Procrustes ANOVA, I complete the classifiers boxes like this : Individual : from dataset Side : none (I work with object asymmetry so I do not have any choice) Error 1 : image Additional main effect : geography. My results are not good, the ANOVA is not the one that I expected. I do not understand the results for the freedom degrees and for F. So I would know if there is a kind of "fluctuating asymmetry analysis step by step" or anything like that which help me to understand the MorphoJ program. Any help is great. Thanks. Marion.
-- *************************************************************** Christian Peter Klingenberg Faculty of Life Sciences The University of Manchester Michael Smith Building Oxford Road Manchester M13 9PT United Kingdom Telephone: +44 161 275 3899 Fax: +44 161 275 5082 E-mail: [email protected] Web: http://www.flywings.org.uk Skype: chris_klingenberg ***************************************************************
