Jay asked:
> Somewhere between the 'simplest' analysis and the most complete
> analysis will lie the one which the analyst can live with, can
> understand, and draw interesting/useful information from. I hope.
> How can one (or we) find a way to provide that level of analysis?
In this medium, by successive approximation, I think. But that method
requires assessing, at each cycle, how closely one has approached the
target. For the present query, we've gone about as far as we can go
with offering advice (which is why we're now discussing what might be
called the <pedagogy of e-mail consultation> rather than the problem
that stimulated the discussion). To assist the querent ("Pingu", aka
"DN") any further, we would need some reaction from him.
E.g., have the several responses so far been:
[check as many as apply: this is NOT a Likert scale! <g>]
__a) helpful
__b) sufficient to perform SOME analysis
__c) enlightening
__d) more informative than one had expected
__e) confusing
__f) incomprehensible
__g) OK so far, but raises questions, viz.:__________________
A few comments about some of the pedagogical issues Jay raises:
On Sat, 12 Jul 2003, Jay Warner wrote [edited]:
> One of the nice things about this type of discussion is that ...
> [one] can be reminded how far an intro stat approach is from what
> _should_ be done.
Or indeed ANY approach.
> A proper full blown consultation for a problem would not jump in as
> I did, but would spend a lot of time exploring the presumptions that
> led to the structure of the data in the first place.
There may be various kinds of "proper". (From one point of view, a
"proper full blown consultation" would require consultation in person,
whence no e-mail exchange could be "proper" in that sense.)
But that quibble aside, there are different ways of approaching any
problem, and their effectiveness depends on how querent and consultant
are severally thinking:
1) Start with the most general situation that is applicable, deal with
it, and then deal with the current problem as a special case;
2) Start with the specific current problem (as specific as perceived
by the consultant, anyway), then extend it as generally as appears to be
useful in the querent's context;
3) Start with whatever is possible actually to DO right now, whether
it be the optimal analysis (e.g.) or not, do it, then contemplate the
outcome(s), looking for evidence that would lead one to want to DO
something else;
and I suppose others can be imagined, depending in part on the nature
of the problem presented in the first place. As Jay writes,
"Action" oriented people see the consulting statistician as an
"abstract thinker."
For such clients, maybe the consultant might adopt approach (3) above,
followed by commentary about why the apparently obvious results might
not mean what one supposes (or would like to suppose (!)) they mean.
> If an advisory suggestion is suited to the level of understanding of
> the analyst, is this acceptable? I fear not.
"Acceptable" surely lies, like beauty, in the eye of the beholder...
How can one tell whether one's "suggestion is suited to the level of
understanding of" the querent, without some response from the querent to
the suggestion? (I don't believe one can, not reliably.)
> Yet simultaneously, adding in the (possible) ramifications and
> implications can tie up said analyst in total knots, unable to move
> intellectually. How to get out of this? Rank the advice in
> ascending order of difficulty? But then the order of assumptions
> required (simplifying assumptions) would be in descending order.
Indeed. As proposed at the beginning of this epistle, "successive
approximation" appears to be the only efficient way of proceeding.
(Indeed, I would argue that this is in fact the method used by any
consultant in face-to-face consultation: it's just that in a
face-to-face situation the cycle time between approximations may
sometimes be measured in minutes (or even fractions of a minute), while
with e-mail the shortest possible cycle time is more likely to be hours
or days. Or weeks, sometimes.)
Well, there's my current $.02-worth on pedagogy...
-- Don.
-----------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
