Ordinal level measures distinguish cases by which characteristic?

Ordinal level measures are designed to categorize and rank cases based on a specific characteristic, allowing researchers to understand the order of cases in terms of their values. This means that ordinal measures can indicate whether one case has more or less of a particular attribute than another, but they do not specify the exact distance between those cases. For instance, in a survey rating satisfaction on a scale from "very unsatisfied" to "very satisfied," the responses can be ranked, indicating that "satisfied" is better than "dissatisfied," without providing a precise quantitative difference between those satisfaction levels.

The distinction of using "greater than" or "less than" is fundamental to ordinal measures because it enables an understanding of the relative position of cases, which is essential for ranking and ordered analysis. This characteristic is what makes the ordinal level distinct from other measurement levels, such as nominal, which only categorizes without ordering, or interval and ratio levels, which include equal intervals and absolute zeros.