ECONOMIC QUALITY CONTROL OF PRODUCT 



379 



pie, provided the variability should be left to chance. Several of the 

 observed values lie outside these limits. This was taken as an indica- 

 tion of the existence of causes of variability which could be found and 

 eliminated. 



Further research was instituted at this point to find these causes of 

 variability. Several were found and after these had been eliminated, 

 another series of observed values gave the results indicated in Fig. 1-b. 

 Here we see that all of the points lie within the limits. We assumed, 

 therefore, upon the basis of this test, that it was not feasible for 

 research to go much further in eliminating causes of variability. 

 Because of the importance of this particular experiment, however, 



48.0 



o 



X 



a> 



a. 

 o 



47.8 



O 



O 

 < 



cr 

 tu 

 > 

 < 



47.6 



L 



4 8 12 16 



SAMPLE 



Fig. 8 — \'ariations that should be left to chance. Does the criterion work? "Yes." 



considerably more work was done, but it failed to reveal causes of 

 variability. Here then is a typical case where the criterion indicates 

 when variability should be left to chance. 



C. Suppose now that we take another illustration where it is reason- 

 able to believe that almost everything humanly possible has been done 

 to remove the assignable causes of variation in a set of data. Perhaps 

 the outstanding series of observations of this type is that given by 

 Millikan in his famous measurement of the charge on an electron. 

 Treating his data in a manner similar to that indicated above, we get 

 the results shown in Fig. 8. All of the points are within the dotted 

 limits. Hence the indication of the test is consistent with the ac- 

 cepted conclusion that those factors which need not be left to chance 

 had been eliminated before this particular set of data were taken. 



