Various Techniques used for testing hypothesis

There are two types of statistical techniques that are used for testing of hypothesis. they are parametric and non-parametric techniques.

• Parametric Techniques -Parametric techniques can be applied for the purpose of testing the hypotheses ie tha following conditions are satisfied.

1. When the sample is randomly selected.
2. When the variances of the various groups are equal or near equal.
3. When the data are in the form of an interval scale or ratio scale.
4. When the observations are independent.
5. When the sample size is more than 30.
6. When the data follow a normal distribution.

• Non-parametric Techniques - When the above conditions are not satisfied, the non-parametric techniques have to be used. The non-parametric tests are population free tests, as they are not based on the characteristics of the population. They do not specify normally distributed populations or equal variances.

The techniques which enable us to compare samples and make inferences or tests of significance without having to assume normality in the populations are known as non-parametric techniques.

Some of the non-parametric techniques are the chi-square test, the rand difference correlation coefficient, the rank difference correlation coefficient, the sign test, the median test, and the sum-parametric tests, that is, they are less able to detect a true difference when such is present. Non-parametric tests should not be used, therefore, when other more exact tests are applicable.

Data have been collected randomly wherein every individual had to respond independently. Since all the conditions requie=red for parametric tests are satisfied, the following techniques are employed.

1. t-test
2. ANOVA
3. w2 (omega square) estimate.