Interesting question. Even when you are interpreting a single data set you probably have some reference data set in mind. If you look at a data set and say to yourself “that data set is very variable (as indicated by its MAD or IQR)” you are implicitly comparing it to a standard data set in your mind that has a smaller “normal” variability. So the point of this standard is to make such comparisons explicit. But that doesn’t mean you would never ask a student to comment on a single data set.