Question: Suppose that someone who works with you performs a hypothesis test to determine whether the mean of a particular process has increased. The sample mean is 50% higher than the null hypothesis mean, but the p-value of the test is 0.18, leading your co-worker to fail to reject the null hypothesis. Is your co-worker making the correct conclusion? If so, how is that possible when the sample mean is 50% higher than the null hypothesis mean? If not, what is she doing wrong?