high quality, but dependable and The above rules and criterion
economic quality, which in turn of meaning were, to him, a nec-
meant quality suited to the pur- essary ingredient of industrial
pose. But what quality is suited research for the reason that, as he
to the purpose? Statistical meth- stated, industrial research is more
ods for discovery of what prod- exacting then pure science. His
uct is needed, what quality is faith in the power of the statisti-
needed, and for learning how a cal method in all human inquiry
product performs in service and was unshakable.
in the laboratory are thus neces- He acknowledged an everlast-
sary ingredients of the statistical ing debt to C. I. Lewis’ Mind and
control of quality. the World Order, which he recom-
The world knows him for the mended to me. I had the usual
Shewhart control charts, and difficulty with it, and I recall say-
the world lives better for them. ing to Shewhart at the end of the
They are, however, only one of seventh reading that so far it had
his statistical contributions. He meant nothing to me. “Stay with
leaves a rich legacy that will take it,” he said. “I read it 14 times
years to absorb. For example, his before it began to mean any-
Rules 1 and 2 on the presenta- thing.” I wonder how he came
tion of data: upon it in the first place, and
Rule 1. Original data should be
how he knew how important it
presented in a way that will pre-
was that he should pursue it.
serve the evidence in the origi-
Although operational defini-
nal data for all the predictions
tions, his criterion of meaning,
assumed to be useful.
and his rules 1 and 2 for the
presentation of data have been
Rule 2. Any summary of a dis-
known to scientists for several
tribution of numbers should not
generations, no one to my knowl-
give an objective degree of belief
edge has stated them so well as
in any one of the inferences or
Shewhart. One sees in them C. I.
predictions to be made there, for
Lewis in the background.
Shewhart’s monumental book, Economic Control of
that would cause human action
significantly different from what
Hypothesis Is Necessary
Quality of Manufactured Product , and the inscription
he wrote in the book he gave his wife: “To the best
this action would be if the origi- Some knowledge must be a pri-
little girl in the world, whose inspiration made this
nal distribution had been taken ori, even if shown later by obser-
book possible 9/31” as a basis for evidence.
vation to be untenable. Without
Then, there is his Criterion of
theory (hypothesis), data are
Meaning:
meaningless or nonexistent.
There is thus no true value of
consumer research, engineer- Every sentence, in
anything; true value is undefin-
ing, or anything else. Otherwise, order to have definite
able [sic] operationally. There are,
the statistician would be merely scientific meaning,
however, numerical values that
another chemist, another agricul- must be practically or
people can use with confidence
tural scientist, or something else. at least theoretically
if they understand their mean-
Quality control meant to him verifiable as either
ing (for the tensile strength of a
[the] use of statistical methods all true or false upon the
batch of wire, for example, or for
the way from raw material to con- basis of experimental
the proportion of the labor force
sumer and back againt— hrough m e a s u r e m e n t s ,
unemployed last month).
redesign of product, reworking of either practically or
There was to Shewhart no
specifications of raw materials— theoretically obtainable
such thing as a random sample.
in a continuous cycle as results by carrying out a definite
There was and is, however, such
come in from consumer research and previously specified
a thing as a sample selected by a
and from other tests. operation in the future.
random operation. There may
He was quick to see that qual- The meaning of such a
be a concept of randomness,
ity must mean not necessarily sentence is the method
but one cannot communicate
of its verification.
20 AMSTAT NEwS SEPTEMbER 2009
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80