1. Regarding the testing of pharmaceutical companies' claims that their drugs are safe, a U.S. Food and Drug Administration official has said that it's "better to turn down 1000 good drugs than to approve one that's unsafe." If the null hypothesis is H0: "The drug is not harmful," what type of error does the official appear to favor?
2. A researcher wants to carry out a hypothesis test involving the mean for a sample of size n 5 18. She does not know the true value of the population standard deviation, but is reasonably sure that the underlying population is approximately normally distributed. Should she use a z-test or a t-test in carrying out the analysis? Why?
3. A research firm claims that 62% of women in the 40-49 age group save in a 401(k) or individual retirement account. If we wished to test whether this percentage could be the same for women in this age group living in New York City and selected a random sample of 300 such individuals from New York, what would be the null and alternative hypotheses? Would the test be a z-test or a t-test? Why?