A friend has performed a significance test of null hypothesis that two means are
ID: 3179129 • Letter: A
Question
A friend has performed a significance test of null hypothesis that two means are equal. His report states that the null hypothesis is rejected in favor of the alternative that first mean is larger than the second. In a presentation on his work, he notes that the first sample mean was larger than the second mean and this is why he chose this particular one-sided alternative. (a) Explain what is wrong with your friend's procedure and why. The first mean can never be larger than the second mean; this indicates a mistake was made during statistical analysis. We should never choose a one-sided alternative. The null hypothesis in this case should have been that the two means were not equal. The null hypothesis in this case should have been that the first mean is larger than the second. We should only choose a one-sided alternative we have some reason to expect a specific directional outcome before looking at the sample results (b) Suppose he reported t = 1.30 with a P-value of 0.11. What is the correct P-value that he should report? ()Explanation / Answer
He has to check the significance test of the null hypothesis that two means are equal.
there is no particular condition to take the one sided alternative hypothesis in the problem
so we should take two sided rather than one sided.
option E is correct.
we should only choose a one sided alternative if we have some particular reason to expect directional outcome before looking at the sample results.
As we have taken two sided then the p value is doubled the one sided p value so p=0.22