I think we are getting close to closure here.
"Irving Scheffe" <[EMAIL PROTECTED]> wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> On Mon, 19 Feb 2001 02:12:46 GMT, "Milo Schield" <[EMAIL PROTECTED]>
> wrote:
> <Snip>
> >But in most of your examples MORE is being claimed. In most cases, the
> >claim includes an inference. Once the claim involves an inference, then
a
> >statistical test may be relevant.
> >
> >In one case, the claim was discrimination (causal explanation of observed
> >differences); in another the claim was greater scoring ability (causal
> >explanation of observed differences).
>
> I don't think so. I think the issue was scoring production,
> not ability, and the example was deliberately restricted:
Scoring production in the past is factual.
Scoring ability is inferential.
There is no argument about the past per se.
There is no issue about what was scored.
The issue/argument is about whether what was observed
will continue -- is a relevant indicator of an underlying
difference that is more than just a coincidence in these particular games.
> I set up the example [basketball scoring] trying to
> [artificially] restrict a discussion which would, naturally,
> evolve in many directions. Of course "point scoring
> ability" is not the same as "points scored," etc.
>
> And, of course, point production is, in the real world,
> not the sole measure of basketball performance.
I've no problem on the measure of performance.
There is no issue/argument about points scored.
The issue is not points scored. That is not arguable.
The issue is productivity -- of which production is a sign.
> >In both cases, the inference involves generalizing from a small "sample"
of
> >time to a larger "population" of time. Thus, the strength of the
argument
> >is influenced by the time-span of the data.
> >
> >How can we measure the influence of the time span involved in the data?
> >Here is where IMHO one can make a case for statistical tests being
RELEVANT.
>
> And?
Jim, I don't know what else to say at this point.
As I see it either you maintain there is nothing inferential in any claim
you've made in any of your argument (in which case I grant you your
conclusion that statsitical tests are irrelevant) or you agree there is an
inferential component to each of your stories -- and that is why people
argue about these matters (in which case statistics may be helpful)..
Either way, this should bring us close to closure.
PS. If discrimination isn't inferential, I don't know what is.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
