Range

How and When to Report the Range

The range belongs to the measures of dispersion and it is used to describe how big the span of all values is. It can be used in cases in which the level of measurements is at least interval-level data. Reporting the range is useful if specific data samples are compared and one wishes to show how much the span of all values differs between the samples. The age example previously described in this entry included a range of 17. If we want to compare this first group to another group of people who are 24, 26, 27, 29, 25, and 29 years old, resulting in a range of 29 − 24 = 5, we can see that the distance between the youngest and the oldest member of both groups differs quite a lot: 17 years compared to 5 years. In that case, reporting the range is of value, especially if one wants to focus on the differences regarding the age span within each groups, for instance, when family structures (e.g., age of siblings and cousins) are described. Imagine our two groups were families; thus, we see that the biggest age difference in the first family is larger than in the second one.

Describing the range for small samples can be useful but, if and how to report it depends on what the researcher wants to describe. On the other hand, the range is of little use in cases in which big data samples are described, because extreme values are still taken into account even if their occurrence is quite infrequent. In the case of such a sample, outliers are specific events that are rare and the question arises of how much value the description of the sample dispersion is if these outliers are taken into account. If these outliers are not only rare but also of extreme values, the range does not mirror the real variability of the sample but blurs it. In that case, range should not be reported. Instead, the inter-quartile range is of more value.

...