Re: The rationale for substantive knowledge of underlying process

[John <jsdwd@ksc.th.com>]

on 5/14/01 8:32 am, Kromkowski@aol.com at Kromkowski@aol.com wrote:
> 
> I said educated 7th graders who are given already available data can create
> and interpret control charts. The operative word is available data.  (Again,
> note the distinction between the construction and interpretation of the
> control chart and actual doing something toward improvement. The rational
> subgrouping argument applies to 1/6 the types of control charts. (C/U/NP/P
> and IndX/Mr do not require subgrouping)  (And the choice of irrational
> subgroup will lead to a special cause.) I'd also like to know how operator

Perhaps more experience with charts would sharpen these discussions.

1.  My experience tells me that X-bar, R charts represent much more than a
sixth of charts in use. (I know of no data to suggest otherwise, but someone
may have done some sort of survey....)  Moreover I think the 'subgroups
argument' (as it is called here) applies to X,Mr charts as well as they can
be used in a variety of ways and even (as Wheeler and others have pointed
out) as  a replacement for attributes charts.  It is not forgone that the
individual points used to make up the chart are successive occurrences.  My
experience also is that C and U charts are encountered infrequently and are
(in any case) not all that useful.  I call them, 'should I be happy or
should I be sad charts'.

2. Why would one have a seventh grader construct a chart that wasn't going
to be interpreted?   The purpose of the chart is to separate special and
common causes and the purpose of doing that is as a guide to action.  If the
action is not possible, the chart is useless (other than as an academic
exercise for 7th graders perhaps).  If a knowledgeable person is needed to
guide the action, I guess the argument that he/she is not needed to actually
construct the chart (an argument with which I don't agree in many cases) is
moot is it not?

John Dowd
jsdwd@ksc.th.com