Population standard deviation is a measure of how spread out values are across an entire group. It tells you, on average, how far each data point sits from the group’s mean. The symbol for it is σ (the lowercase Greek letter sigma), and it’s one of the most fundamental concepts in statistics.
If you have test scores for every student in a school, σ tells you whether most students scored close to the average or whether scores were scattered widely. A small σ means the values cluster tightly around the mean. A large σ means they’re spread far apart.
How It Differs From Sample Standard Deviation
The word “population” is doing important work here. In statistics, a population is every member of the group you care about: every student in a school, every lightbulb produced in a factory run, every temperature reading from a full year. When you have data on all of them, you’re working with population data, and you use σ.
Most of the time, though, you don’t have data on everyone. You have a sample, a smaller subset drawn from a larger group. Sample standard deviation uses the symbol s and adjusts its calculation slightly to account for the fact that a sample tends to underestimate the true spread. The adjustment is small but meaningful: instead of dividing by the total number of data points (N), the sample version divides by N minus 1. This correction produces a more accurate estimate of the real population’s variability.
The rule of thumb: if your dataset includes 100% of the group you’re analyzing, divide by N. If it’s a subset of a larger group, divide by N minus 1. The rest of the math is identical.
The Five Calculation Steps
Computing population standard deviation by hand is straightforward. Here’s how it works, using a small example: suppose you’re looking at the ages of all five employees at a company: 25, 30, 35, 40, 45.
- Step 1: Find the mean. Add all the values and divide by the number of data points. Here, (25 + 30 + 35 + 40 + 45) ÷ 5 = 35.
- Step 2: Find each distance from the mean, then square it. For each data point, subtract the mean and square the result. So: (25 − 35)² = 100, (30 − 35)² = 25, (35 − 35)² = 0, (40 − 35)² = 25, (45 − 35)² = 100.
- Step 3: Add up all the squared distances. 100 + 25 + 0 + 25 + 100 = 250.
- Step 4: Divide by N, the total number of data points. 250 ÷ 5 = 50. This result is called the variance (σ²).
- Step 5: Take the square root. √50 ≈ 7.07. That’s your population standard deviation.
The squaring in step 2 exists for a reason. Without it, the positive and negative distances from the mean would cancel each other out and sum to zero every time. Squaring makes all values positive. The square root in the final step brings the result back into the original units, so if you’re measuring ages in years, your standard deviation is also in years.
How Standard Deviation Relates to Variance
Variance and standard deviation measure the same thing, just on different scales. Variance (σ²) is the average of the squared distances from the mean. Standard deviation (σ) is the square root of variance. Variance is useful in many statistical formulas because squared values are easier to manipulate mathematically, but it’s harder to interpret on its own because the units are squared. If your data is in dollars, variance is in “dollars squared,” which isn’t intuitive. Standard deviation brings the number back to dollars, making it far more practical for everyday interpretation.
What the Number Actually Tells You
A standard deviation is most useful when you pair it with the mean. Saying a group of students has an average test score of 75 with a standard deviation of 5 paints a very different picture than an average of 75 with a standard deviation of 20. The first class performed consistently. The second had huge variation between the highest and lowest scorers.
When data follows a normal distribution (the familiar bell curve), standard deviation maps onto specific percentages that are worth memorizing. About 68% of all values fall within one standard deviation of the mean. About 95% fall within two standard deviations. And about 99.7% fall within three. This is known as the empirical rule.
So if the average height of adults in a country is 170 cm with a population standard deviation of 7 cm, roughly 68% of adults are between 163 cm and 177 cm. About 95% are between 156 cm and 184 cm. Only about 0.3% fall more than three standard deviations from the mean, beyond 149 cm or 191 cm. Those individuals are genuine statistical outliers.
When You’d Actually Use It
Population standard deviation applies whenever you have complete data on the group you’re analyzing. National census data is one classic example: if a government records the income of every household in the country, calculating σ from that dataset gives you the true population standard deviation, not an estimate. Complete manufacturing data is another common case. If a factory tests every single unit in a production batch, the variation across those units is a population measure.
Classroom settings often use population standard deviation too. If a teacher wants to know how spread out the scores were on a particular exam, and the dataset includes every student who took it, that’s the full population. There’s no larger group to estimate.
In practice, true population data is relatively rare. Researchers almost always work with samples and use the sample standard deviation (with the N minus 1 correction) to estimate the population’s spread. But understanding the population version first makes the sample version easier to grasp, which is why statistics courses teach it in this order.
Notation Conventions
Statistics uses a consistent system to distinguish population values from sample values. Population measures, called parameters, use Greek letters: σ for standard deviation, μ (mu) for the mean, σ² for variance. Sample measures, called statistics, use Latin letters: s for standard deviation, x̄ (x-bar) for the mean. If you see σ in a formula or textbook, the author is referring to the population. If you see s, they’re referring to a sample.
Calculating It in Software
You’ll rarely compute standard deviation by hand outside of a classroom. In Excel or Google Sheets, the function STDEV.P calculates population standard deviation. The “P” stands for population. The sample version is STDEV.S, or simply STDEV. Getting these mixed up is one of the most common spreadsheet mistakes in statistics.
In Python’s NumPy library, the standard deviation function defaults to population standard deviation (with degrees of freedom set to zero). To get the sample version instead, you set the degrees of freedom parameter to 1. This default catches some users off guard, so it’s worth checking which version your tool assumes before trusting the output.