Suppose that we wish to test the Null Hypothesis Ho that the proportion p of ledger sheets of a company that are incorrect is = 1/20 versus Ha: p > 1/20.
Suppose we decide to use the following test procedure: 2 ledger sheets are selected at random and if both are incorrect, we reject the null hypothesis. If one or more of them is (are) correct, we look at a 3rd ledger sheet. If this one is incorrect, we reject the null hypothesis (otherwise, we fail to reject the null hypothesis)
For this test, can you find the type-I error alpha? Compute the power of this test at the alternative value p = 1/10