CLASS-L Archives

May 2005


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
"F. James Rohlf" <[log in to unmask]>
Reply To:
Classification, clustering, and phylogeny estimation
Wed, 25 May 2005 22:23:00 -0400
text/plain (64 lines)
There are many methods that have been proposed. The simplest is to express
them as ratios of some measure of size - which gets back to the original
question as there are many ways to express size. The sum of all the
measurements is a pretty unusual choice but it often may not matter much in
practice because the measurements are usually very highly correlated.

F. James Rohlf, Distinguished Professor & Graduate Program Director
State University of New York, Stony Brook, NY 11794-5245

> -----Original Message-----
> From: Classification, clustering, and phylogeny estimation 
> [mailto:[log in to unmask]] On Behalf Of leo horseman
> Sent: Wednesday, May 25, 2005 10:03 PM
> To: [log in to unmask]
> Subject: 
> I do not have a citation for you; however, I am interested in 
> the question.  
> I must admit to being baffled.  Why include size measurements 
> in your analysis at all, unless somehow the measurement of 
> shape is derived from the size measurement.  If you've 
> measured only bone lengths, which surely would vary greatly 
> in size, then how did you arrive at a shape measurement?
> M.C.
> >From: Richard Wright <[log in to unmask]>
> >Reply-To: "Classification, clustering, and phylogeny 
> estimation"            
> >   <[log in to unmask]>
> >To: [log in to unmask]
> >Date: Mon, 23 May 2005 12:06:54 +1000
> >
> >This is a question about size and shape in morphometric studies.
> >
> >I have a series of measurements of bone lengths in individuals that 
> >vary greatly in size. I want to consider similarities and 
> differences 
> >between the individuals solely in terms of shape. To remove 
> differences 
> >in absolute size I have taken each individual, summed all the 
> >measurements for that individual, and then divided each 
> measurement by that sum.
> >
> >This is an intuitively simple method. Some years ago I read a 
> >recommendation about using it. Now an editor is asking me to cite a 
> >reference to the use of the method. Unfortunately I cannot recollect 
> >where I saw it advocated.
> >
> >Can anybody help?
> >
> >I know that there are various alternative methods for 
> eliminating size 
> >in multivariate morphometric work, such as eliminating the first 
> >principal component if that is one of general size. However 
> my question 
> >is not about the competing merits of size/shape methodologies in 
> >general. This is a specific request for a citation of the 
> approach outlined above.
> >