Is it correct that Spearman's cannot be used when data is at the ordinal level?

Spearman's rank correlation coefficient is specifically designed to assess the strength and direction of the association between two ranked variables, which can include ordinal data. It is a non-parametric measure, meaning it doesn't assume any specific underlying distribution of the data, such as normality. This flexibility allows researchers to utilize Spearman's correlation with ordinal data without any constraints based on distribution characteristics.

Given that ordinal data inherently involves ranking, using Spearman's correlation can provide meaningful insights even when the data don't meet the stricter requirements of parametric tests (which often presume interval or ratio-level data and normal distribution). Thus, it is indeed correct that Spearman's can be applied effectively to ordinal data, contributing to the rationale for choosing the answer indicating that it is false to assert that Spearman's cannot be used in such cases.