What Is Standard Deviation?

Standard deviation measures how spread out a set of numbers is. It answers a simple question: are the values in this data set bunched up close together, or are they scattered all over?

If the standard deviation is small, the values are tightly packed around the average. If it is large, the values reach far from the average in both directions.

For a fast number on any list of values, the standard deviation calculator takes a comma-separated, space-separated, or line-separated list and reports the mean, the standard deviation, and the variance in one step.

In one sentence: standard deviation is the typical distance between any value in a data set and the mean of that data set.

Imagine you measured how late your friends arrived at a dinner. You worked out the average lateness. Standard deviation tells you the typical gap between any single arrival and that average. If standard deviation is two minutes, most arrivals were within a couple of minutes of the average. If it is fifteen minutes, you have at least one very early or very late friend pulling the spread wide.

The full mathematical definition involves squaring distances and taking square roots, but the underlying idea is just “average distance from the mean.”

Two data sets can share the same mean and look very different once you look at standard deviation:

• Set A: 4, 5, 5, 5, 6. The mean is 5. The values barely move from 5. Standard deviation is small.
• Set B: 1, 2, 5, 8, 9. The mean is also 5. The values fan out wide. Standard deviation is large.

The mean alone cannot tell those two stories apart. Standard deviation is the second number that fills in the gap.

Low and high are best understood by comparison.

• Low standard deviation. Values cluster close to the mean. The data is consistent and predictable. Test scores in a class where most students landed within a few points of the average have low standard deviation.
• High standard deviation. Values fan out from the mean. The data is variable. Daily returns on a volatile stock, or weights of a mixed crowd of athletes, tend to have high standard deviation.

Whether high or low is “good” depends entirely on the situation. Low standard deviation is a positive signal for things you want to be reliable (manufacturing tolerances, test scoring, savings account growth) and a negative one for things you want variety in.

Take this small data set: 2, 4, 4, 4, 5, 5, 7, 9.

Step 1. Find the mean.

(2 + 4 + 4 + 4 + 5 + 5 + 7 + 9) / 8 = 40 / 8 = 5

The mean is 5.

Step 2. For each value, find the distance from the mean and square it.

• (2 − 5)² = 9
• (4 − 5)² = 1
• (4 − 5)² = 1
• (4 − 5)² = 1
• (5 − 5)² = 0
• (5 − 5)² = 0
• (7 − 5)² = 4
• (9 − 5)² = 16

Step 3. Average the squared distances. That number is the variance.

(9 + 1 + 1 + 1 + 0 + 0 + 4 + 16) / 8 = 32 / 8 = 4

Variance is 4.

Step 4. Take the square root.

√4 = 2

Standard deviation is 2. So in this data set, values land an average of 2 units from the mean of 5.

(This example uses the population formula, dividing by the count. Sample standard deviation divides by count minus one instead. Most data analysis tools default to the sample version.)

The mean is the anchor that standard deviation measures distance from. Without a mean, there is nothing to compare values to.

That is why every standard deviation calculation has a mean in the middle of it: you need the mean before you can ask “how far is each value from this average?”

One small consequence: if every value in your data set is the same number, the mean equals that number, every distance is zero, and standard deviation is zero. Zero spread is possible; negative spread is not.

Variance and standard deviation are two ways of describing the same idea: how spread out a data set is. They are connected by a simple rule:

Standard deviation = √variance

Variance is the average of the squared distances from the mean. Standard deviation is just the square root of that number, which puts the answer back into the same units as your data. If your data is in dollars, variance is in squared dollars (which nobody can intuitively read), and standard deviation is in dollars (which everyone can).

For a calculator that reports both at the same time, see the variance calculator. It also breaks out population vs sample variants. For a full side-by-side comparison of the two measures and when to use each, see Variance vs Standard Deviation.

Standard deviation shows up across statistics, finance, science, and quality control because the mean alone hides the story.

• Test scores. Two classes can share an average of 80, but if one has standard deviation 3 and the other has 18, the second classroom looks completely different.
• Investment returns. Two funds can have the same average annual return, but the one with the larger standard deviation swings harder from year to year.
• Manufacturing. A part has a target dimension, and a tight standard deviation across a production run means the line is consistent.
• Forecasting. Many error bars and prediction intervals you see on charts are built from standard deviations.

In every case, the same number (standard deviation) is doing the same job: describing how much the data wiggles around its average.

A few practical rules of thumb make standard deviation easier to read:

• Compare standard deviation to the mean. A standard deviation of 2 around a mean of 100 is small. A standard deviation of 2 around a mean of 4 is large.
• For data that is roughly bell-shaped, about 68 percent of values fall within one standard deviation of the mean, and about 95 percent fall within two. This is a useful shortcut, but it only applies to data that follows a normal distribution.
• Standard deviation is sensitive to outliers because it squares distances. One unusually high or low value can pull standard deviation up sharply, even if most of the data is clustered.
• Standard deviation cannot be negative. The smallest possible value is zero, which only happens when every value in the data set is the same.

Treat the number as a description of consistency, not as a verdict. Whether a particular standard deviation is good or bad depends on what the data is meant to represent.

Doing the math by hand is a useful exercise once. After that, a calculator is faster and less error-prone. Five steps and one square root for every data set adds up.

The calculator approach:

• Paste your numbers as commas, spaces, or one per line.
• Read the mean, both standard deviation variants (population and sample), and both variances at once.
• Move on with your day.

For more than a handful of values, the standard deviation calculator saves a lot of arithmetic. Keep the manual method in your back pocket for when you want to double-check what the calculator is doing. For the formulas in detail and a step-by-step walkthrough of the math, see Standard Deviation Formula.

• Standard deviation measures how spread out a data set is.
• Low standard deviation means values cluster near the mean.
• High standard deviation means values fan out from the mean.
• It is calculated using each value's distance from the mean.
• Standard deviation = √variance. They describe the same spread in different units.
• Standard deviation is never negative.
• The standard deviation calculator gives you the answer in seconds.