Question 8.17

8.17 What's wrong

Explain what is wrong with each of the following.

  1. The large-sample confidence interval for a population proportion is based on a t statistic.
  2. A large-sample significance test for an unknown proportion is plus or minus its standard error.
  3. You can use a significance test to evaluate the hypothesis versus the two-sided alternative.

8.17

(a) It is based on a statistic, not . (b) margin of error, not standard error. (c) does not belong in the hypotheses; it should be .