One should not use the Spearman r on skewed interval data, unless the data could be converted into ordinal form.

Question

Quizplus · Accepted Answer

The Spearman r is a nonparametric correlation coefficient used for ordinal data. If interval data is skewed, it cannot be treated as ordinal data and therefore Spearman r should not be used. It is better to use other nonparametric tests like Kendall’s tau or a parametric test like Pearson's correlation coefficient after transforming the data if necessary.

One Should Not Use the Spearman R on Skewed Interval