Minitab uses either the binomial or the hypergeometric distribution to calculate acceptance probabilities for go/no go attribute data.

By default, Minitab uses the binomial distribution to find and compare sampling plans. To use the binomial distribution appropriately, you must be able to assume that your sample comes from a large lot (at least 10 times the sample size) or that your sample comes from sampling a stream of parts selected at random from an ongoing process. For the grocery bags data, 25,000 is a large lot; it is much larger than 10 times the sample size of 306 bags.

If your data are sampled from isolated, finite lots, you can choose to have Minitab use the hypergeometric distribution to calculate the probability of acceptance. This option is only available when using go/no go data.

If the grocery store was trying a new bag vendor who gave you an experimental lot of 1,000 bags, you might want to specify the hypergeometric distribution.