This document summarizes parametric and nonparametric tests. Parametric tests make assumptions about the population based on known parameters, while nonparametric tests make no assumptions about the population. Some examples of parametric tests provided are t-test, F-test, z-test, and ANOVA, while examples of nonparametric tests include Mann-Whitney, rank sum test, and Kruskal-Wallis test. The key differences between parametric and nonparametric tests are that parametric tests are based on population parameters and distributions while nonparametric tests are not, and parametric tests can only be applied to variable data while nonparametric tests can be used for variable or attribute data.
2. ∗ If the information about the population is completely
known by means of its parameters then statistical test is
called parametric test
∗ Eg: t- test, f-test, z-test, ANOVA
Parametric Test
3. ∗ If there is no knowledge about the population or
paramters, but still it is required to test the hypothesis of
the population. Then it is called non-parametric test
∗ Eg: mann-Whitney, rank sum test, Kruskal-Wallis test
Nonparametric test
5. Difference between parametric and Non
parametric
Parametric Non Parametric
Information about population is
completely known
No information about the population is
available
Specific assumptions are made regarding
the population
No assumptions are made regarding the
population
Null hypothesis is made on parameters of
the population distribution
The null hypothesis is free from
parameters
6. Difference between parametric and Non
parametric
Parametric Non Parametric
Test statistic is based on the distribution Test statistic is arbritary
Parametric tests are applicable only for
variable
It is applied both variable and artributes
No parametric test excist for Norminal
scale data
Non parametric test do exist for norminal
and ordinal scale data
Parametric test is powerful, if it exist It is not so powerful like parametric test
7.
8. ∗ Non parametric test are simple and easy to understand
∗ It will not involve complecated sampling theory
∗ No assumption is made regarding the parent population
∗ This method is only available for norminal scale data
∗ This method are easy applicable for artribute dates.
Advantages of non parametric test
9. ∗ it can be applied only for norminal or ordinal scale
∗ For any problem, if any parametric test exist it is highly
powerful.
∗ Nonparametric methods are not so efficient as of
parametric test
∗ No nonparametric test available for testing the interaction
in analysis of variance model.
Disadvantages of non parametric test