In: Statistics and Probability
Does it make sense to report a confidence interval if the primary goal is to perform a hypothesis test? Why or why not?
It makes sense to report a confidence interval if the primary goal is to perform a hypothesis test because confidence interval provides more information including the hypothesis test. Hypothesistesting is a binary decision: whether torejectthe null hypothesis or fail to reject the null hypothesis. Confidence interval provides a range of values in which the true unknown population parameter lies. From the confidence interval, we can always deduce the information required by hypothesis test.
Thus, since confidence interval covers the outcomes of hypothesis testing, it makes sense to report a confidence interval if the primary goal is to perform a hypothesis test