Question

Use the concept of practical significance to explain why a confidence interval is (usually) preferable to a hypothesis test.

Answer 1

Confidence interval is an estimate of a range where we think the parameter values lies.

We can use confidence interval (CI) for hypothesis testing. In general, if the CI for an effect does not span 0 then the null hypothesis can be rejected. But a CI can be used for more, whereas reporting whether it has been passed is the limit of the usefulness of a test.

The reason CI is used instead of just a t-test, for example, is because then more can be one than just test hypotheses. we can make a statement about the range of effects to be likely (the ones in the CI). we can't do that with just a t-test. we can also use it to make statements about the null, which we can't do with a t-test. If the t-test doesn't reject the null then we just say that can't reject the null, which isn't concluding much.