On Thu, 24 Aug 2000, Ken Reed wrote in part:
KR> > One variable has 4 categories (agree-neutral-disagree, don't know).
> >
DB> Are you trying to say that you have one such variable, and your other
> > variables are otherwise described; or that you have a number of such
> > variables and you want something like an item analysis of them all; or
> > that you have a number of such variables that you intend to combine in
> > some unspecified way to produce the variable you want to test; or ...?
>
KR> My question is about just the one variable.
> >
DB> By the way, this variable is not one variable, it is two: (1) degree
> > of agreement with whatever, and (2) whether the respondent has an
> > opinion about it. If you have a bunch of variables like this, what
> > you can do with them depends partly on how much missing data (=
> > "don't know" responses) you have.
KR> "don't know" is a meaningful response in this case -- it is not
> 'missing' data.
For the variable "degree of agreement with proposition X", the response
"don't know", however meaningful it may be otherwise, _is_ missing data.
The respondent claims not to know something: whether s/he doesn't know
what "proposition X" means, or doesn't know what his/her opinion would be
on that proposition if s/he were more familiar with it, or has some other
internal reaction to the question, _you_ do not know this person's
opinion, nor do you know what this person's opinion would be if whatever
impediment exists (to reporting an opinion) were removed.
I will concede that all this is largely irrelevant if you decide to
pursue an analysis of the 4-valued nominal variable, ignoring the ordinal
relationship that unambiguously holds among 3 of those values (and the
interval relationship that could probably be argued -- that the distance
between "agree" and "neutral", and the distance between "neutral" and
"disagree", are not detectably different).
-----------------8X----------------------------------------------
KR> > 2. Is there a rule-of-thumb for what would be evidence of strong
> > > within-group agreement?
> >
DB> Rules of thumb exist only to help one avoid having to think hard
> > about some situation or problem. As such they are invariably heavily
> > dependent on contexts, about which we have very little information.
>
KR> Perhaps. But they also signal conventions.
Yes. The root of "convention" and "convenience" is the same (convenire,
to come together), and the convenience of a convention lies precisely in
not having to think about the reason(s) for the rule one is invoking, so
long as the conditions are met under which the convention applies. But
that requires knowing SOMEthing about what those conditions are, and how
they are met in the present circumstances.
-- Don.
------------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED]
MSC #29, Plymouth, NH 03264 603-535-2597
184 Nashua Road, Bedford, NH 03110 603-471-7128
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
