Standard Deviation Calculator

The key difference lies in the denominator: sample standard deviation divides by (n-1) while population standard deviation divides by n.

Use sample standard deviation when your data is a subset of a larger population (which is most common in real-world scenarios).

Use population standard deviation only when you have data for every member of the entire population you're studying.

For population: σ = √[Σ(xi - μ)² / N], where μ is the population mean and N is the population size.

For sample: s = √[Σ(xi - x̄)² / (n-1)], where x̄ is the sample mean and n is the sample size.

The division by (n-1) in the sample formula is called Bessel's correction and provides an unbiased estimate of the population standard deviation.

Standard deviation helps you understand data variability.

In a normal distribution, about 68% of values fall within one standard deviation of the mean, 95% within two standard deviations, and 99.7% within three standard deviations (the 68-95-99.7 rule).

Compare standard deviations across different data sets to understand which has more variability.

Standard deviation is a measure of how spread out a set of numbers is around its mean (average). It answers the question: on average, how far does each data point sit from the center?

A small standard deviation means the values cluster tightly around the mean, while a large one means they are widely scattered.

According to Encyclopaedia Britannica, standard deviation is the most widely used measure of dispersion in statistics because it is expressed in the same units as the original data, unlike variance. It underpins everything from quality control to finance, letting you compare consistency across different data sets at a glance.

To calculate standard deviation by hand, follow five steps:

• First, find the mean by adding all values and dividing by the count, which is exactly what our average calculator does in one click if you want to check that figure first.
• Second, subtract the mean from each value to get the deviations.
• Third, square each deviation so negatives do not cancel positives.
• Fourth, add the squared deviations and divide by n for a population or n-1 for a sample to get the variance.
• Fifth, take the square root of the variance.

Khan Academy walks through this same procedure with worked examples. This calculator automates each step and shows the intermediate values so you can check your own work.

Consider the classic data set 2, 4, 4, 4, 5, 5, 7, 9. The sum is 40 across 8 values, so the mean is 5.

The deviations from the mean are -3, -1, -1, -1, 0, 0, 2, 4, and their squares are 9, 1, 1, 1, 0, 0, 4, 16, which add up to 32.

For a population, divide by N = 8 to get a variance of 4, and the square root gives a population standard deviation of exactly 2. This example appears in Wolfram MathWorld's standard deviation entry.

For a sample, you would divide 32 by n-1 = 7, giving a variance of about 4.571 and a sample standard deviation of about 2.138.

The population standard deviation formula is sigma = sqrt(sum of (xi - mu)^2 / N), where mu is the population mean and N is the number of values.

The sample formula is s = sqrt(sum of (xi - xbar)^2 / (n-1)), where xbar is the sample mean. The only difference is the denominator.

The NIST/SEMATECH e-Handbook of Statistical Methods defines both and explains that dividing by n-1, known as Bessel's correction, yields an unbiased estimate of the population variance from a sample.

Variance is simply the value under the square root, so standard deviation is always the square root of variance and shares the original data's units.

Sample standard deviation divides the sum of squared deviations by n-1 rather than n because a sample tends to underestimate the true spread of the whole population.

When you compute deviations from the sample mean instead of the unknown population mean, you lose one degree of freedom, so the divisor is reduced by one to compensate. This adjustment, called Bessel's correction, produces an unbiased estimator of the population variance.

The NIST e-Handbook confirms n-1 is the standard choice for sample data. Use n-1 whenever your numbers are a subset drawn from a larger group, which covers most surveys, experiments, and real-world data collection.

Standard deviation appears across many fields:

• In finance, it measures the volatility of an investment's returns, so a higher standard deviation signals higher risk.
• In manufacturing and quality control, it drives Six Sigma processes that keep products within tolerance.
• In education, it describes how tightly test scores cluster around the class average.
• In weather and climate science, it captures temperature variability across seasons.

Encyclopaedia Britannica notes that standard deviation is central to the normal distribution, which models countless natural phenomena. Because it is expressed in the same units as the data, professionals use it to compare consistency, set thresholds, and flag unusual observations.

For data that follows a normal (bell-shaped) distribution, the empirical rule states that about 68% of values fall within one standard deviation of the mean, roughly 95% within two standard deviations, and about 99.7% within three.

So if test scores have a mean of 70 and a standard deviation of 10, about 68% of students score between 60 and 80.

Wolfram MathWorld describes this pattern as a defining property of the normal distribution. The rule turns an abstract spread number into a concrete probability, helping you judge how common or rare a given value is within your data set.

• The most frequent error is mixing up the sample and population formulas by using the wrong denominator; dividing by n when you should use n-1 understates the spread for sample data.
• Another mistake is forgetting to square the deviations, which lets positive and negative differences cancel out to zero.
• People also confuse variance with standard deviation, reporting the squared value instead of its square root.
• Rounding the mean too early propagates errors through every later step, so keep full precision until the end.
• Finally, standard deviation only summarizes symmetric, roughly normal data well; for heavily skewed data it can mislead, as Khan Academy cautions.