-------- Original Message -------- Subject: Re: Modularity in the human skull Date: Tue, 5 Apr 2011 13:59:25 -0400 From: Chris Klingenberg <[email protected]> Reply-To: [email protected] Organization: University of Manchester To: [email protected] Dear Hans The test in the Procrustes ANOVA is for differences in the means. Computing a pooled within-group covariance matrix (i.e. pooled within-sex in your case) removes just that difference in means. Best wishes, Chris On 4/5/2011 6:18 PM, morphmet wrote:
-------- Original Message -------- Subject: Modularity in the human skull Date: Sun, 3 Apr 2011 07:06:05 -0400 From: Hans Wellens <[email protected]> To: <[email protected]> Dear All, To investigate modularity in the modern human skull, I`ve collected a sample of some 200 subjects (mixed male and female, no specific age restrictions). The idea is to calculate the multi-set RV coefficient (Klingenberg) for three different scenarios often hypothesized to exist by orthodontists and maxillofacial surgeons. MorphoJ would be used for this purpose. I first wanted to make sure that male and female subject could effectively be pooled. The idea was to determine whether they differed in shape (differences in size are to be expected) and whether the underlying covariance structures were similar. The Procrustes ANOVA, which is based on the covariance matrix of a pooled generalized Procrustes superimposition, indicates a highly significant difference in shape. When superimposing the male and female mean configurations, the differences are however incredibly small (hardly visible). I can of course run the modularity tests separately for both sexes, but that renders basically the same results as running the test on the pooled sample. Nevertheless they seem to vary slightly differently. On the other hand, it would seem logical to assume the same, or at least highly similar interactions, be it anatomic, genetic, or functional, should occur between the modules of male and female patients. So my common sense tells me to go ahead and pool the sample, notwithstanding the result of the Procrustes ANOVA. What do you think? Hans Wellens, DDS Orthodontist Groene-Poortdreef 16 8200 Sint-Michiels, Brugge [email protected]
-- *************************************************************** 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 ***************************************************************
