Question

1. Upon which type of data could parametric tests be applied? Upon which type of data could nonparametric tests be applied?

2. What are at least five assumptions that must be met or nearly met to use parametric tests?

Answer 1

Parametric tests can be applied to the type of data that is anticipated to have been drawn from the known distribution. E.g. Students T-Test.

For interval scales or ratio scales, use parametric statistics.

Non parametric tests can be applied when we don’t assume population parameters. E.g. chi-Square Test

For nominal scales or ordinal scales, use non parametric statistics.

Assumptions about the underlying data to perform parametric statistics are:

1.       Interval or continuous level of values/measurements.

2.       Samples are random.

3.       Independent samples.

4.       Samples come from the population that is normally distributed.

5.       Population variances are equal.