Question

Q81 Explain the advantages and disadvantages of nonparametric versus parametric testing. [4 Marks]

DO NOT WRITE THE ANSWER - USE WORD FORMAT.

NO PLAGIARISM ACCEPTED IN THE ANSWER.

Answer 1

Advantages of Non-Parametric Test over Parametric Test:

1. Non parametric test are readily comprehensible, very simple and easy to apply. And always applicable with some weak conditions. But in parametric test if sample size is small and population distribution not normal then parametric tests fail.
2. In Non parametric test we don’t need to know the parent population from which sampling is done. But in parametric test parent population must be normal population.
3. For testing the randomness of sample, we have test for randomness based on runs in non-parametric test but parametric test does not exist for testing the randomness of the sample.
4. Non parametric test are robust but parametric test is not robust.
5. Non parametric exist to deal with mere classification (i.e. which are measured in nominal scale) while Parametric test does not exist for such data.

Disadvantages of Non-Parametric Test over Parametric Test:

1. Non parametric tests can be applied only for nominal or ordinal measurements. Non parametric test (if exist) is more powerful.
2. No Non-parametric test exist for testing interaction in ‘ Analysis of Variance’ model unless special assumptions about the additively of the model are made.
3. Non parametric test is designed to test statistical hypothesis only and not for estimating the parameters.