Discuss the below:
Why do you think there is such a controversy between one-sided and two-sided p- values?
Why can a small mean difference in a t-test produce a large p-value when the sample size is small and a small p-value if the sample size is large?
If you know that a 90% confidence interval for the difference of two means does not include 0, what can you say about the p-value for testing µ1= µ2 vs. µ1=-µ2?