QUALITY CONTROL CHARTS 



595 



controlled during the period for which the data have been collected. 

 In this instance, however, it was possible to group the 15,050 obser- 

 vations into twelve groups representing monthly samples of approx- 

 imately 1250 instruments each. The data are presented in this form 



SEP TEMBER 



Quality. Defined by Characteristic 

 X 



Fig. 2 — Monthly polygons showing distribution in quality for samples of approxim- 

 ately 1250 units of product. Do these data present any ex'idence of lack of control? 



in Fig. 2. Obviously no two polygons are the same in respect to aver- 

 age, dispersion and shape, but of course we would not expect them to 

 be the same even though the product were uniform, any more than we 

 would expect two targets to show the same distribution of shots even 



