Consider the large-sample level .01 test in Section 8.4 for testing H0: p=.2 against Ha: p>.2. a) For the alternative value p=.21, compute beta(.21) for sample sizes...

Consider the large-sample level .01 test in Section 8.4 for testing H0: p=.2 against Ha: p>.2.

a) For the alternative value p=.21, compute beta(.21) for sample sizes n=100,2500,10,000,40,000,and 90,000.

b) For p^=x/n=.21, compute the P-value when n=100, 2500, 10,000, and 40,000.

c) In most situations, would it be reasonable to use a level .01 test in conjunction with a sample size of 40,000? Why or why not?