nominal characters

Mr Fortuner connection modem fortuner at MATH.U-BORDEAUX.FR
Sat Apr 8 08:37:13 CDT 1995

```The original question about intraspecific variability (ISV) of nominal
characters has shifted to theoretical considerations about various
systematics/ecological/genetic/consumer behavior/etc. issues. This was to be
expected because any lively discussion will branch out and spread before it
dies.

On the other hand, I am still interested in one very precise question
about one very precise type of character. I should have started by redefining
the 4 types of characters (ratio/interval/ordinal/nominal) but I thought
everybody knew that (I found it in a 1975 book by Pankhurst - chapter written
by Rypka who quotes an earlier work by Andersen). These definitions have now
been quoted by Warren Lamboy. Thanks.

I did add to these definitions by
differentiating between those nominal characters that can be represented by
one or several ordered characters and those that cannot.

For example, color
per se obviously is a nominal character (is red bigger than yellow?). However,
any color can be described in terms of wave length, grey level and chroma
(saturation). (There may be other values to consider, such as luminance which
is measured in nits, but let's not go into that). Wave length and grey level
are ratios, chroma is an interval (the 14 steps of Munsell). Intraspecific
variability (be it from genetic or environmental origin) is not surprising in
a ratio character such as wavelength, so color itself should be expected to
vary as well. In the same manner, if seedlessness can be linked to a ratio
character (e.g., the length of the seed), then ISV is not surprising.

What I
am interested in are nominal characters that CANNOT be represented by ordered
(=ordinal, interval, or ratio) characters. Obviously, I am talking of a
relation between actual states, not ordering of an arbitrary code (0/1)
attached to these states.

My questions are:

1) Are there any such
characters?
2) If one can be found, does it have ISV?

But as I said, this
discussion has gone to other issues (which I find very interesting by the
way). So, don't mind me, go ahead, but if anyone comes across an answer to my
questions, I would appreciate hearing about it.

Renaud
Fortuner
fortuner at math.u-bordeaux.fr

Ohmegosh! We are having a discussion on
a DISCUSSION network! This cannot be!... Quick, where is my unsubscribe
button?

```