Level of Measurement Which of the levels of measurement (nominal, ordinal, interval, ratio) describe data that cannot be used with the methods of rank correlation? Explain.

Data are measured on different levels of measurement.

The measurement levels are divided into four categories:

Nominal: Data that contains categories/labels of a particular variable is measured on a nominal scale.

Ordinal: Data that contains categories but can be arranged in order is measured on an ordinal scale.

Interval: Numerical data that can be arranged in an order,where the distance between successive values is known is measured on an interval scale.

Ratio: Numerical data that contains a well-defined natural zero starting point and whose ratios can be computed is measured on the ratio scale.

For data measured on thefirst two levels of measurement, namely nominal and ordinal, the differences between the observations are not defined.

Ranks deal with observations that are ordered, and the difference between any twosuccessive ranks is equal.

Since the differences between the sample values are not defined, ranks cannot be assigned to such observations.

Therefore, the rank correlation method cannot be applied for data measured on the nominal scale and the ordinal scale.