Whether you have ordinal or interval data depends on the nature of
the relationship between your measurements and the true scores (positive
monotonic versus positive linear). Since the true scores can never be known
in this life, your question falls in the domain of metaphysics, not
statistics. You can avoid the issue, in part, by defining the truth on your
measurements -- that is, you decide to generalize your results to that
reality in which the constructs you have measured are linear transformations
of the measurements you have at hand. Or do you think that there is a
single concrete reality apart from your existence and that it can be known?
I have no problem using ordinal data with parametric analysis, as
long as other assumptions, most notably the usual normality assumption, are
not seriously violated. The classic nonparametric analyses are little or
nothing more than the classic parametric analyses done after transforming
the data to ranks.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karl L. Wuensch, Department of Psychology,
East Carolina University, Greenville NC 27858-4353
Voice: 252-328-4102 Fax: 252-328-6283
[EMAIL PROTECTED]
http://core.ecu.edu/psyc/wuenschk/klw.htm
-----Original Message-----
In the medical research I review, the outcome variables are often rating
scales. Examples would be a visual analogue scale for pain or
rating scales (usually from 0 - 10) of various other factors. These
data are often compared using parametric statistics. My question
is.....are these scales interval/ratio or are they ordinal. My feeling is
that these are ordinal data unless you can show me studies that prove the an
8 is truly twice as much as a 4. Therefore I have always used
non-parametric statistics. Am I correct????
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
