Question

In which case would testing the null hypothesis involve a two-tailed statistical test?

Answer 1

We use the two-tailed statistical test, when the alternative hypothesis contains the '' sign. In this case we have to test for both the tails. Example if we want to test that mean content of a particular quantity is equal to 200g or not. Then our hypothesis would be:

H₀: v/s

H_a: .

If the level of significance () = 0.05, then the critical value would be at 0.025 level of significance (), that is the rejection region would be on both tails rather than just one tail.