Hello Jonathan,

The form you propose: Mdn(Q1-Q3) is fine; as well as presenting a 50% highest density region, the values can be used for crude inferences about skew or symmetry in a distribution. I'd label the values parenthetically the first time if you're presenting a series of such values, or as a table note if in tabular form. Yes, you would include a decimal value where appropriate.

As well, Tukey's 5-number summary can be informative: Min score, Lower hinge, Median, Upper hinge, Max score. Tukey's hinges are close to, but not necessarily identical to, Q1 and Q3 as percentile points.

Good luck with your work.

Table is a good idea and at the same time necessary. If the numbers with decimal figured is respectively low (for example less than 10 percent), rounding is a good idea as well.

The idea of professor Morse to report five numbers is better and more informative. I do it in my works because it gives a bare minimum scheme about the distribution of the data of interest.