What type of data does a parametric test typically require?

Parametric tests require data that meets certain assumptions, one of which is that the data is measured on an interval or ratio scale. These types of data possess meaningful distances between values, allowing for the calculation of means, standard deviations, and other statistical measures that parametric tests often rely on.

Interval data has equal distances between points but does not have a true zero point, while ratio data includes a true zero, enabling all mathematical operations. Because parametric tests, such as t-tests and ANOVAs, rely on these properties to make inferences about populations based on sample data, interval and ratio data are essential for their validity. This requirement distinguishes parametric tests from non-parametric tests, which can handle nominal and ordinal data but may be less powerful in detecting true effects when the assumptions of parametric tests are met.