Learn
Interactive·

Spread: Range, Variance & Standard Deviation

The same mean can hide completely different levels of variation.

Suppose two classes both have an average score of 70. In one class, most scores may fall between 68 and 72; in the other, they may range from 40 to 100. The means are identical, but the learning gaps are not.

Spread describes how far values lie from the center. A measure of center tells you where the data sit; a measure of spread tells you how widely they are dispersed around that point.

Key question

Which number should summarize how far the observations are spread around the center?

1

Concept

What does spread describe?

Measures of spread summarize variability in different ways

MeasureFormula or definitionFeature
Rangemaximum − minimumsimple, but uses only the endpoints
IQRQ3 − Q1width of the middle 50%
Variancesum of squared deviations divided according to sample sizeuses every observation
Standard deviationsquare root of variancesame unit as the original data
2

Why It Matters

Is the range enough?

The range is easy to compute but depends on only two values

The data sets 2, 4, 6, 8, 10 and 2, 2, 6, 10, 10 both have a range of 8. Yet the first is evenly spread, while the second piles up at the endpoints. The range cannot show that difference.

Effect of an extreme value

If the maximum changes from 10 to 100, the range jumps from 8 to 98. The IQR may move much less because it focuses on the middle half of the data.

3

How It Works

How are variance and standard deviation built?

Standard deviation summarizes spread around the mean in the original unit

First subtract the mean from each observation to get deviations. Raw deviations cancel because some are positive and others negative, so we square them and add them. Divide by n for a population or usually by n−1 for a sample to obtain variance, then take the square root to return to the original unit.

Sample variance and sample standard deviation

s² = Σ(xᵢ − x̄)² / (n−1), s = √s²

Why divide by n−1?

The sample mean is estimated from the same data, which uses one degree of freedom. Dividing by n−1 reduces the tendency to underestimate the population variance.

4

Example

Which measure of spread should you use?

The distribution shape guides the choice of spread measure

Data situationRecommended summary
Roughly symmetric with few extremesMean and standard deviation
Skewed or containing extremesMedian and IQR
Full width mattersAlso report the range
Comparing different unitsConsider a standardized measure such as the coefficient of variation

Reading the standard deviation

A standard deviation of 10 points does not mean that every value is exactly 10 points from the mean. It summarizes the overall scale of dispersion around the mean.

5

Interactive

Change the spread

Change the spread

Keep the mean fixed at 10 while moving the values farther from the center. Watch how the range, variance, and standard deviation increase even though the mean does not change.

Range

8.0

Sample variance

10.00

Standard deviation

3.16

6.0
8.0
10.0
12.0
14.0

What to notice

All measures of spread grow as observations move farther apart. Variance responds especially strongly because it squares the distances.

Key takeaways

  • Spread describes how widely values are dispersed around the center.
  • The range is simple but highly dependent on the minimum and maximum.
  • Variance uses squared units; standard deviation returns to the original unit.
  • Mean and standard deviation often suit symmetric data, while median and IQR often suit skewed data.
Next: Percentiles & Quartiles