YOUR TURN
~ 7 min reading + activities
I well remember that, the first few times I was preparing to run the Red Beads Experiment in public, I felt rather nervous. Wouldn’t you—knowing that you (as the Foreman) were going to be spending an hour or more just talking about 24 as-yet-unknown random numbers (or as near random as makes no difference)? But I soon found out that it could hardly be easier! Whatever numbers turn up, it’s so simple to find “reasons” for them. Of course it is: we are well-practised at the art. We do it every week, month, quarter, half-year, year, in our management reports, with production figures, sales figures, financial figures. Financial commentators do it every day—or even every hour or less, e.g. regarding the movements of stock market indices and exchange rates. I’m not saying that all the reasons reported are wrong. I am pointing out that, whatever are the results, we and they come up with reasons—whether there actually are any or not! Even on a day when, say, the stock market index hardly moves, a reason is given for that! Few bosses or customers are yet ready to hear explanations just versed in terms of common-cause variation. Therefore they are given special-cause explanations all the time, for every result. What clearer contradiction in terms could there be? Yes, of course the Foreman praises a lower number of defectives. Of course he criticises a larger number. Of course he remarks on consistent performance. Of course he remarks on erratic performance. Of course he complains if the target isn’t being met. Of course!
Now comes the time for you to get some practice at playing the role of the Foreman! Remember that the Foreman (pretends that he) is in the Un-Knowing state—typified by all the discussion in the so-called “Analysis” of the results on pages 12 and 13, starting in A Brief Overview. So, to help you get ready to perform in that role, I recommend that you now go back and re-read that page and a half before continuing here.
After a couple of pages of introduction to the Experiment on Red Beads at the beginning of DemDim Chapter 6, the data presented in Activity 2–f are discussed from the top of DemDim page 103. So now that you have worked your way through this Activity, I suggest you complete this morning’s study by reading Chapter 6 (it’s only eight pages long). This will both consolidate what we have already covered and also perhaps provide you with a few extra points to add to the notes that you are taking to help you with the Major Activity at the end of the day. (As usual, ignore the formulas and computations if you are on Stats-level 0.)
Reference is made during Chapter 6 to the Funnel Experiment since, in DemDim, that is covered in the previous chapter. Here we shall be catching up with the Funnel Experiment tomorrow. In fact, in tomorrow’s Major Activity, which will occupy almost the entire afternoon, you will be carrying out a full version of the Funnel Experiment (although using different equipment from that described by Dr Deming).
ACTIVITY 2–f
The next four pages show you, one at a time, the results from the Red Beads Experiment illustrated in DemDim Chapter 6.
There is space for you to insert indications of your reaction to the number of red beads obtained at each stage. Naturally, you could say even more if the experiment were “live”: you could see things to congratulate or criticise about the way the Willing Workers are carrying out their task (wrong angle, or too fast, or too slow, or too unsteady … ). But there’s still plenty that can be said in response to the figures alone!
To get you started, I have included my suggestions of reactions to the first four figures—they are very much the kind of comments that you might have heard from Dr Deming himself. See how you get on with the rest of the experiment. If, after some thought, you need extra help then I’ve given a few further suggestions on page 27, below.
NB The version of the experiment in DemDim Chapter 6 is Dr Deming’s. I should therefore mention three minor differences between his version here and my version which is otherwise being illustrated today:
Describe this table
A table of red-bead counts with a row for each of the six workers (Audrey, John, Al, Carol, Ben and Ed) and a Daily Totals row beneath them, and a column for each of Days 1 to 4 plus a Totals column. At this first step, only the very first cell — Audrey, Day 1 — is filled in: a count of 16. Every other cell is empty. Each successive table on this page adds one new count, building up the experiment value by value. The pedagogical point is that the Foreman reacts to each new number as it lands, attributing it to the worker, the day, the training, the appraisal — even though, taken as a whole series, these numbers are pure common-cause variation around a stable process.
“What a terrible start! But you’ve only just been trained—weren’t you watching?”
“That’s better.”
“Excellent: continuous improvement.”
“What a disappointment.”
Here are a few more suggestions:
Now, if you haven’t done so already, go back and try to complete your remarks for all the remaining results.
(Then, finally, take a look at my complete set of remarks on Appendix page 8.)
The Foreman’s Transcript
Here is every reaction you gave—to pure, random variation. Read them back and let the absurdity sink in.