From the list given to you for each of the following hypothesis testing situatio
ID: 3172972 • Letter: F
Question
From the list given to you for each of the following hypothesis testing situations, indicate the non-parametric test you would use and explain the reason
1) One Sample Sign Test (Test # 6)
2) Wilcoxon Signed-Rank Test (Test# 7)
3) Mann-Whitney (U) Test: Two Independent Samples (Test# 8)
a) Researchers want to assess the efficacy of a drug for managing patients with moderate hypertension. They recruit 60 patients and randomly assigned them to two groups. Group A will receive Drug A; and group B will receive Drug B.The investigators want to assess whether there is a difference in the blood pressure readings after the participants have received a 30-day supply of the specified medication. Ho: Drug A and Drug B perform the same in lowering blood pressure. Data is not normally distributed.
b) A researcher wants to assess if the median income of people who voted for President Trump is greater than the median income of people who voted for Hillary Clinton.
c) A researcher is interested in assessing whether a smoking cessation seminar is effective in reducing smoking. Forty participants are selected for the study. Daily average smoking figures for the week before and week after the seminar are recorded. The researchers find that the differences in computed scores are substantially skewed.
Explanation / Answer
a) This is the case of Mann whitney U test , which is the non parameteric variation of the independent samples t test. here we have independent samples of 2 groups and we are given that data is NOT normally distributed and we want to compare the difference in the blood pressure reading of the 2 groups. hence we perform a MWU test
b) We perform a one sample sign test to determine whether the median of a group differs from a specified value/other group. Here we want to compare the median of 1 group with the median of the other . Hence we shall conduct a 1 sample sign test
c) The wilcox test is the non parametric alternative of the dependent t test. here we have a dependent group as we have a "BEFORE" and "AFTER" kind of set up. So as the data is not normal , we go for wilcox test here .
Hope this helps
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.