LEARN MORE...

Home
Documents
February 2012
3
Res 342 Week 3 DQ 1

Res 342 Week 3 DQ 1

Content type

User Generated

Type

Homework

Uploaded By

flrqfununo

Pages

Showing Page:

1/1

Week 3 DQ 1: Under what circumstances must a nonparametric test be used?

Explain. What are the strengths and weaknesses of nonparametric tests? Can the

outcomes of nonparametric tests be generalized to populations?

A non-parametric or distribution-free test must be used when the focus of the research

is on the sign or rank of the data rather than the exact numerical value of the variable, not to

specify the shape of the parent population, often can be used in smaller samples and used for

ordinal data in which the measurement scale is not interval or ratio. I can recall from the

previous course of one measurement scale that did not get mention here which is nominal. In

addition, non-parametric tests cannot be used by a parametric data.

There are always strengths and weaknesses in tests regardless of what type. The

strengths or advantages of non-parametric tests include that the tests can be used in small

samples, generally more powerful than parametric tests when normally cannot be assumed,

and can be used for ordinal data, distribution free, with the small samples researchers can

compute the exact P value from the sample unlike the parametric test.

On the other hand, weaknesses or disadvantages of non-parametric tests include

require special tables for small samples and if can be assumed, parametric tests are generally

more powerful. Moreover, the non-parametric tests cannot be presented by a parametric model

and when these assumptions are satisfied, they are not as good as the parametric tests.

Whether parametric on nonparametric, when researchers define a population and

conduct a test with a random sample, the outcome of the tests apply to the entire population.

Therefore, the answer is yes, the outcomes of nonparametric tests be generalized to

populations.

In general, because nonparametric tests make fewer assumptions about the population,

rejection of a hypothesis using nonparametric tests is convincing. The text mentions the use of

nonparametric test in business world in various departments include human resources and

marketing. However, in accounting and finance, management may encounter skewed

populations that render parametric tests unreliable. In this scenario, management or analysts

would use nonparametric tests as a compliment to their customary parametric tests. I would be

curious to know the need in using both, parametric and nonparametric tests to prove a

hypothesis.

References

Doane, D. & Seward, L. (2007). Chapter 16: Applied statistics in business and economics. Burr

Ridge, Illinois: The McGraw-Hill.

Unformatted Attachment Preview

Week 3 DQ 1: Under what circumstances must a nonparametric test be used? Explain. What are the strengths and weaknesses of nonparametric tests? Can the outcomes of nonparametric tests be generalized to populations? A non-parametric or distribution-free test must be used when the focus of the resea ...
Purchase document to see full attachment