Standard Deviation Calculator

Standard deviation is the most widely used measure of how spread out a set of numbers is. Two classes can both average a 75% on a test, yet one class might have scores tightly bunched between 70 and 80 while the other ranges from 40 to 100. The averages hide that difference; the standard deviation reveals it. This calculator takes any list of numbers and instantly returns the sample and population standard deviation, both variances, the mean, and a step-by-step breakdown so you can see exactly where each figure comes from.

As a quick worked example, take the data set 10, 12, 23, 23, 16, 23, 21, 16. The mean is 18, the population standard deviation is about 4.90, and the sample standard deviation is about 5.24. The sample figure is slightly larger because it divides by N − 1 instead of N - more on that below.

The standard deviation formula

Both formulas start the same way - by measuring how far each value sits from the mean, squaring those distances, and adding them up. The only difference is what you divide by:

Here x is each value, μ (or x̄) is the mean, N is the number of values, and Σ means "sum of." Variance is the same expression without the square root - so standard deviation is just the square root of the variance, and both are reported by the calculator.

Sample vs. population: which one do I use?

This is the question that trips people up most. The rule is simple:

• Population (divide by N): use this when your data includes every member of the group you care about - for example, the test scores of all 8 students in a class when those 8 students are the entire population of interest.
• Sample (divide by N − 1): use this when your data is a subset drawn from a larger group, and you want to estimate the spread of that larger group - for example, 8 randomly chosen voters used to estimate a whole electorate.

When in doubt, most statistics courses and research default to the sample formula, because real data is almost always a sample of something bigger. That is why this calculator opens in sample mode.

Why divide by N − 1? (Bessel's correction)

When you estimate a population's spread using only a sample, you have to use the sample mean rather than the true population mean. The sample mean always sits as close as possible to the sample's own values, which makes the deviations a touch too small and biases the variance downward. Dividing by N − 1 instead of N nudges the result upward just enough to remove that bias. This adjustment is called Bessel's correction, and its effect shrinks as your sample grows: with 5 values the correction is large, but with 5,000 it is negligible.

How to use this calculator

1. Enter your numbers in the box, separated by commas, spaces, or new lines. You can paste a column straight from a spreadsheet.
2. Choose the data type - Sample (N − 1) or Population (N) - using the toggle. The big headline result switches to match your choice.
3. Read the headline figure at the top: that is the standard deviation for the mode you selected, with the variance just beneath it.
4. Scan the statistics grid for the mean, median, sum, count, range, and both versions of the standard deviation and variance side by side.
5. Follow the step-by-step and the squared-deviation table to see exactly how the answer was built - useful for homework you need to show work for.

The result updates instantly as you type, and any non-numeric text you accidentally include is ignored and flagged.

Who this calculator is for

• Students checking statistics homework and needing to show each step of the working.
• Teachers generating worked examples or verifying answer keys.
• Researchers and analysts who want a fast spread figure without opening a spreadsheet.
• Quality and process engineers tracking how consistent a measurement or output is.
• Investors gauging the volatility (risk) of a series of returns.
• Anyone who needs to know whether a set of numbers is consistent or all over the place.

Worked example 1: a small data set by hand

Take the classic set 2, 4, 4, 4, 5, 5, 7, 9. The sum is 40 and there are 8 values, so the mean is 5. The deviations from the mean are −3, −1, −1, −1, 0, 0, 2, 4; squaring them gives 9, 1, 1, 1, 0, 0, 4, 16, which sum to 32. For the population, variance = 32 ÷ 8 = 4, so σ = √4 = 2. For the sample, variance = 32 ÷ 7 ≈ 4.571, so s ≈ 2.138. Notice the sample figure is a little larger - exactly what Bessel's correction does.

Worked example 2: comparing two data sets

Imagine two employees' weekly sales. Alex sells 48, 52, 50, 49, 51 units; Sam sells 20, 80, 35, 70, 45. Both average 50 units a week. But Alex's sample standard deviation is about 1.58, while Sam's is about 24.7. Same average, wildly different consistency - the standard deviation is what tells you Alex is the steadier performer, a distinction the mean alone completely misses.

Worked example 3: spotting an outlier's effect

Start with 10, 11, 12, 13, 14 (mean 12, sample SD ≈ 1.58). Now swap the last value for an outlier: 10, 11, 12, 13, 114. The mean jumps to 32 and the sample standard deviation explodes to roughly 45.9. Because deviations are squared, a single far-off value has an outsized impact on the standard deviation - a key reason to inspect your data for typos and genuine outliers before trusting the figure.

The empirical rule (68-95-99.7)

For data shaped like a normal "bell curve," standard deviation has a handy interpretation. The reference table shows what share of values fall within a given number of standard deviations of the mean:

So in normally distributed data, a value more than two standard deviations from the mean is uncommon (about 1 in 22), and one beyond three standard deviations is rare (about 1 in 370). To express exactly how many standard deviations any single value sits from the mean, convert it to a z-score with the Z-Score Calculator. The rule is an approximation and only holds for roughly bell-shaped data - skewed data needs other tools.

Key terms explained

• Mean (x̄): the arithmetic average - the sum of all values divided by how many there are.
• Deviation: how far a single value lies from the mean (x − x̄). Deviations always sum to zero, which is why we square them.
• Variance: the average of the squared deviations. It is in squared units.
• Standard deviation: the square root of the variance, back in the original units.
• Range: the gap between the largest and smallest value - a crude spread measure that ignores everything in between.
• Coefficient of variation: the standard deviation as a percentage of the mean, useful for comparing data with different scales.

Standard deviation vs. other spread measures

Standard deviation is not the only way to describe spread, but it is usually the most useful. The range is simple but depends entirely on the two most extreme values. The interquartile range is robust to outliers but harder to compute by hand. Mean absolute deviation averages the unsigned distances from the mean and is more intuitive, but standard deviation's squaring step makes it mathematically convenient and the foundation of most statistical methods - which is why it dominates in practice.

Practical tips

• Match the formula to the question. If you are describing the data you have in full, use population; if you are inferring something about a bigger group, use sample.
• Keep the units in mind. Report standard deviation in the same units as your data and variance in squared units, and never mix them up.
• Compare alongside the mean. A standard deviation of 5 is small for incomes but huge for shoe sizes - context is everything.
• Use the coefficient of variation when comparing data sets measured on different scales.
• Always eyeball the data first. A quick scan catches the outliers and typos that can wreck a standard deviation.

Related statistics calculators

This page answers "how spread out is my data?" For neighboring questions, a sister tool fits better:

• For the mean, median, and mode on their own, use the Average Calculator.
• To see how many standard deviations a single value sits from the mean, use the Z-Score Calculator.
• To turn a spread into the chance of an outcome, use the Probability Calculator.
• For percentages and percent change, use the Percentage Calculator.
• For grade point averages, use the GPA Calculator.
• For ratios and proportions, use the Ratio Calculator; for very large or very small numbers, use the Scientific Calculator.

Standard deviation in the real world

The reason standard deviation shows up everywhere is that "spread" is a universal question. A few of the fields that lean on it daily:

• Finance and investing: standard deviation of returns is the standard measure of volatility, and therefore risk. A fund averaging 8% a year with an SD of 4% is far steadier than one averaging 8% with an SD of 20%, even though their returns match. Risk-adjusted measures such as the Sharpe ratio divide return by standard deviation to compare them fairly.
• Manufacturing and quality control: "six sigma" literally refers to keeping defects more than six standard deviations from the mean. Control charts flag a process the moment a reading drifts beyond two or three SDs of the target, catching problems before they become scrap.
• Education and testing: standardized exams report scores in standard deviations from the mean. An SAT or IQ result is often expressed as a z-score, which you can compute from the SD using the Z-Score Calculator.
• Science and medicine: error bars on a chart are usually one standard deviation (or a standard error derived from it), and clinical reference ranges are commonly set at the mean plus or minus two standard deviations of a healthy population.
• Weather and sports: "above-average rainfall" or a "career-best season" only carries meaning relative to the typical year-to-year spread, which is exactly what the standard deviation captures.

Standard deviation vs. standard error

These two terms are constantly confused, but they answer different questions. Standard deviation describes the spread of the individual data points you have - how scattered the actual numbers are. Standard error (more precisely, the standard error of the mean) describes how precise your estimate of the mean is, and it equals the standard deviation divided by the square root of the sample size. As you collect more data, the standard deviation stays roughly the same - it is a property of the data - but the standard error shrinks, because a larger sample pins down the average more tightly. When a report shows error bars, check the caption: bars of one standard deviation tell you about variability in the data, while bars of one standard error tell you about confidence in the average. This calculator reports the standard deviation; you can derive the standard error by dividing the sample SD by the square root of the count.

A quick note on rounding

Round only at the very end. If you round the mean before computing deviations, the small error is then squared and summed across every value, and it can noticeably shift the final standard deviation - especially with a small data set or a mean that is not a whole number. This calculator carries full precision internally and rounds only the displayed figures, so the step-by-step table may show neat rounded numbers while the headline result reflects the exact calculation. When you check work by hand, keep several decimal places on the mean and the squared deviations and round the standard deviation last.

Sources

• NIST/SEMATECH e-Handbook of Statistical Methods - measures of scale (standard deviation, variance, range).
• NIST/SEMATECH e-Handbook of Statistical Methods - the normal distribution and the empirical 68-95-99.7 rule.
• Definitions follow standard textbook statistics, with sample figures using Bessel's correction (division by N − 1) for an unbiased estimate of population variance.