Author Question: When should you use a nonparametric test? If the normality assumptions are met, are you better off ... (Read 78 times)

pepyto · **on:** Jun 24, 2018

When should you use a nonparametric test? If the normality assumptions are met, are you better off using a parametric t-test to a nonparametric Rank Test?

What will be an ideal response?

Question 2

Find a 90 confidence interval for the mean rating given by all freshmen and sophomores on the campus.

A) 3.98 0.592
B) 3.98 0.229
C) 3.98 0.712
D) 3.98 0.456

shoemake · **Reply #1 on:** Jun 24, 2018

Answer to Question 1

Frequently, the assumption of normality is reasonable. Moreover, by virtue of the central limit theorem, many of these test procedures remain approximately valid when applied to large samples even if the population distribution is not normal. However, it may be the case that in practical applications the normality assumption is not tenable. In these circumstances, it is desirable to base inferences on nonparametric tests that are valid over a wide range of distributions of the parent population. Such tests are often referred to as distribution-free tests. The parametric tests tend to use all the sample information. Hence, they do not lose the information that is lost when ranking the data. Intuitively they are better, because you know more. Mathematically they are better because they are more powerful. If a parametric test is reasonable it is preferable to a rank test.

Answer to Question 2

B