RE: graphing Ketan's weight

[Steven_S_Prevette@rl.gov]

>With all due respect to those who have tried to analyze Ketan's data, I
beg to differ. It shouldn't be done.

Data are still data, and there is information to be gained from good
analysis of data.  It is amazing how many times I am brought a batch of data
after the fact and told to figure out what happened.  Certainly it is less
than an ideal situation from an analyst's point of view, but still is
something that needs to be done.  Did any changes made make a difference?
The control chart can at least point out to us if anything changed in the
data.  Usually in the cases where folks were not monitoring the data during
the change, the data turn out to be stable, and the change was ineffective.

Now certainly in Ketan's data there are some flaws.  The worst is that the
data are not at all obtained on anything near to a consistent interval.  We
have to guess what happened during the long intervals with no weight
recorded.  But we can still take a start and see what can be gained from the
data, and then try to work with Ketan to improve his future data gathering,
and even have an influence on his future ability to lose weight, assuming
that is the aim.  We have more hope on being a positive influence on Ketan's
system if we work with what we have, rather than sending him away.

Steve Prevette
Site Technical Authority for Statistical Trending
Environment, Safety and Health
Fluor Hanford, A Fluor Global Services Company
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
509-373-9371