That choice is "unusual" because most workers do not use that method. One
reason is probably that one would not measure the size of a box by adding
its length and width - most would rather multiply them together.
The problem with "size" is that simple word hides a large diversity of
possible measures. One cannot decide what is size without specifying some
model and what desired properties the measure of size is supposed to
optimize. One should not use some measure just because it seems "reasonable"
intuitively. As I said previously, it is fortunate that in most biological
applications the various choices are usually highly correlated.
-----------------------
F. James Rohlf, Distinguished Professor & Graduate Program Director
State University of New York, Stony Brook, NY 11794-5245
www: http://life.bio.sunysb.edu/ee/rohlf
> -----Original Message-----
> From: Classification, clustering, and phylogeny estimation
> [mailto:[log in to unmask]] On Behalf Of Richard Wright
> Sent: Thursday, May 26, 2005 1:52 AM
> To: [log in to unmask]
> Subject: Re: size
>
> I should have written: "I have a series of linear
> measurements of bones . . ." They in fact include several
> measurements of breadth and thickness.
>
> I wonder why the sum of all measurements should be such an
> unusual choice.
>
> Richard Wright
>
> >
> >Subject: Re: size
> > From: "F. James Rohlf" <[log in to unmask]>
> > Date: Wed, 25 May 2005 22:23:00 -0400
> > To: [log in to unmask]
> >
> >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
> >www: http://life.bio.sunysb.edu/ee/rohlf
> >
> >> -----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.
> >> >
> >>
>
|
