for =a
Fred Barney & Joe McCollum
For several generations we have seen those who have to decide on the quality of a given shipment lot use tables that with the specification of a few parameters allow the user to choose the desired sample size and after inspection reach a conclusion about the quality of the shipment. For this purpose these tables are fine, if you can convince yourself that the data in question is Normally distributed and you need not be clear why the recommendation of the table on acceptability should be taken. Alternatively, these tables are find if you are willing to sample with replacement and your strongest basis for accepting the judgement contained in the tables is that other people are doing it.
The author's contribution to this area is the discovery of a statistics one book and a computer.
for =a
for b
With the above formulation of the acceptance problem and the hither to undiscovered model of sampling without replacement, the hypergeometric model, the authors have written a computer program that solves the same problem as the high price consultants that translate those incomprehensible tables from the precomputor era