Re: Special Causes Indicators in Red Bead

[Paul Hollingworth <PH@paulholl.demon.co.uk>]

Most people learn best with practical examples. The red bead experiment
is the most robust table top experiment I have ever encountered, but I
often get some bizarre patterns. 

One memorable occasion was in Moscow in 1993 when we were down to the
last four workers. Two of them were very powerful & influential business
men (ex KGB / 'party officials', though I did not know it at the time)
their last two draws were: worker A 2,2 worker B 17,19. A real zig-zag!
Lots of fun to be had with red beads in Russia, all common cause of
course ;-)

My all-time favourite was a particularly enterprising young lady who put
her paddle into the beads upside down. Of course she drew 0 red beads
(and no whites either!). The inspectors went along with this (as they
will often do when trickery is afoot). Though I saw what they were
doing, I chose to ignore it at the time. When later I sacked all the
workers I explained that one of the complaints we had received from the
customer was that we were shipping empty boxes! It showed up as a
special cause, of course.

Don't forget: Walter Shewhart showed that the fact that a point is
outside the control limits does not mean that there has to be an
assignable cause for it, only that it is economically worth
investigating as such. 

Don Wheeler's expressions "possible signals" & "probable noise" are
lucid.


PH

----------------------------------------------------------------
Paul Hollingworth                       4GM Consulting  
email: PH@4GM.com                       http://www.4GM.com
phone: +44(0) 1423 322225               fax: +44(0) 1423 322205 
----------------------------------------------------------------