The range measures the full span of a data set by subtracting its smallest value from its largest. It offers a quick view of spread, though one extreme value can shift it sharply. That makes it simple to calculate—and worth comparing with measures such as variance and standard deviation.

Multiple Choice

What is the dispersion measure defined as the difference between the largest and smallest value in a data set?

This is about how spread out the data are. The quantity defined as the difference between the largest and smallest value in a data set is the range. It gives a simple, overall sense of the dispersion by showing the full span of the data from the minimum to the maximum. Variance and standard deviation, by contrast, measure how far individual values tend to deviate from the center (the mean) and involve squaring those deviations, which makes the measure more sensitive to outliers and more computationally involved. The mean, meanwhile, is a measure of central tendency, not spread. So the range best fits the idea of dispersion described.

The Range: A Quick Way to See How Spread Out Data Are

A set of numbers can have a clear center and still vary widely. Imagine two classes with the same average score: in one, most students’ scores sit close together; in the other, they stretch from very low to very high. The average alone won’t show that difference. A measure of spread can—and one of the simplest is the range.

The range is the difference between the largest and smallest values in a data set. It gives a quick snapshot of the full span of the data, from one end to the other. If a group of temperatures runs from 12°C to 23°C, for instance, its range is 23 − 12, or 11°C.

That calculation is easy. Interpreting what it tells us takes a little more care.

Find the two endpoints

To calculate the range, identify the minimum and maximum values, then subtract the minimum from the maximum:

Range = maximum value − minimum value

Consider the numbers 4, 6, 8, 10, and 12. The smallest value is 4, and the largest is 12, so the range is 12 − 4 = 8.

The values in between matter less for this calculation. Whether the middle numbers are 6, 8, and 10 or 5, 5, and 11, the range stays the same as long as the endpoints remain 4 and 12. That’s useful when someone wants a fast summary. It’s also a clue to the range’s main limitation: it tells us nothing about how the values are arranged between the endpoints.

For example, the sets 1, 2, 3, 4, 10 and 1, 5, 6, 9, 10 both have a range of 9. Yet their values are distributed differently. The first set has four values close to the low end, while the second is more balanced across the span. The range sees only the boundaries, not the full picture.

A wide span—or just one unusual value?

The range gives a sense of the overall spread, but it can be strongly affected by an outlier: a value that sits far from most of the others.

Suppose a small shop records the number of customers who visit each day: 41, 44, 45, 46, and 78. The range is 78 − 41 = 37. That single busy day stretches the range considerably. Most of the other days are clustered between 41 and 46, but the range doesn’t reveal that pattern.

This sensitivity isn’t always a flaw. If the aim is to know the full span of observed values—for example, the highest and lowest temperatures recorded in a month—the endpoints may be exactly what matters. But if the aim is to describe what is typical, a range can paint a distorted picture when an unusual observation appears.

A useful habit is to look at the data themselves before drawing a conclusion. The range is a summary, not a full story. Like checking the widest gap between two fence posts, it tells us how far apart the edges are, but not whether the ground between them is smooth, crowded, or uneven.

How the range differs from variance and standard deviation

The range is one of several ways to describe dispersion, or how spread out values are. Two other common measures are variance and standard deviation. These use every value in the data set, rather than just the two endpoints.

Both measures begin with the mean, or average. They look at how far each value lies from that center. Variance squares those differences and then averages them. Standard deviation is the square root of the variance, which brings the result back into the same units as the original data.

That sounds more involved than finding the range because it is. But the extra work can reveal patterns the range misses. If values tend to sit close to the mean, the standard deviation will generally be smaller. If many values lie farther away, it will usually be larger.

Each measure answers a slightly different question:

  • Range: How far apart are the smallest and largest values?

  • Variance: How much do values vary around the mean, measured using squared deviations?

  • Standard deviation: How far do values tend to lie from the mean, in the data’s original units?

The range can be a handy first glance, while variance and standard deviation offer a broader view of the data’s spread. One isn’t automatically better in every situation. The useful measure depends on what someone needs to understand.

Units make a difference

Because the range is found by subtracting one value from another, it uses the same units as the original data. A range of 11°C stays in degrees Celsius; a range of 8 centimetres stays in centimetres.

This can make the result easy to interpret. If the shortest plant in a greenhouse is 14 cm tall and the tallest is 31 cm, the range is 17 cm. That tells us the total distance between the shortest and tallest recorded plants.

But units can also limit comparisons. A range of 17 cm might be large for seedlings and modest for trees. Even when the units match, different groups may contain very different numbers of observations, or have very different typical values. So a range should be read in context, not as a stand-alone verdict.

For a rough comparison across groups, it helps to know the scale of the measurements and what the endpoints represent. A range of five dollars means something different for a cup of coffee than for a monthly utility bill.

Small data sets and missing context

The range is often most informative when a data set is small and its values are easy to inspect. With only a handful of observations, the minimum and maximum can offer a quick sense of the spread. As a data set grows, though, those two values may become less representative of the group as a whole.

There’s another detail worth checking: how the data were collected. A range depends on the observations included. If a sensor missed a very hot afternoon, the recorded temperature range may be narrower than the true span. If someone enters a value incorrectly, that error could create a misleading maximum or minimum.

In other words, the arithmetic is simple, but the data still deserve a quick look. Ask whether the endpoints are plausible and whether they reflect the full group you care about. This is not a fancy statistical ritual; it’s the same common sense used when a single surprising number catches your eye in a household budget.

What does the range have to do with probability?

Descriptive statistics summarize data that have been observed. Probability, by contrast, describes how likely different outcomes are under a model or set of assumptions. The two areas often meet.

Imagine a weather service estimating tomorrow’s temperature as likely to fall between 16°C and 22°C. That interval has a span of six degrees. It resembles a range, though a probability forecast may involve more information than the observed minimum and maximum in a data set. The forecast might assign different chances to different temperatures, while a simple range treats every value between its endpoints alike.

That distinction matters. A range tells us the width of a span; it doesn’t tell us how likely any value inside that span is. Nor does it promise that future observations will stay within the same limits. If historical daily temperatures ranged from 10°C to 28°C, that record describes the values already seen, not every value that could occur.

So the range can offer a useful first connection between data and uncertainty, but it shouldn’t be mistaken for a probability statement. To understand likelihood, we need information about possible outcomes and their chances—not just the lowest and highest observations.

A simple measure with a clear job

The range has an appealing kind of honesty: it answers one narrow question directly. Subtract the smallest value from the largest, and you have the total observed span.

That makes it useful for quick summaries, basic comparisons, and situations where the endpoints matter. It also means we shouldn’t ask it to do more than it can. The range doesn’t show where most values sit, how unevenly they’re spread, or how likely future outcomes may be.

Used alongside other summaries—and with a glance at the data themselves—the range is a good starting point. It gives the edges of the picture. For the details inside those edges, we need a closer look.