How this calculator works
Standard deviation is a single number that summarises how spread out a list of values is around its mean. Low standard deviation means the values cluster tightly around the mean; high standard deviation means they are scattered. It is the most used measure of spread because it is in the same units as the data (dollars, kilograms, degrees).
The calculation has three steps. First, find the mean: add every value and divide by how many there are. Second, for each value, subtract the mean and square the result — squaring removes the sign so positive and negative deviations contribute equally. Third, add those squared deviations and divide by either n (population) or n − 1 (sample) to get the variance; the standard deviation is the square root of the variance.
The population formula divides by n and gives σ (sigma). Use it when your list contains every member of the group you care about — the heights of everyone on a specific sports team, or the grades of all 30 students in a class if the class is the whole population.
The sample formula divides by n − 1 and gives s. Use it when the list is a sample drawn from a larger group and you want to estimate the population standard deviation — a poll of 1,000 voters, a batch of 50 tested components. n − 1 is Bessel's correction and prevents the sample from systematically underestimating the real spread.
The empirical rule for roughly normal distributions says about 68% of values fall within one standard deviation of the mean, 95% within two, and 99.7% within three. This is a shorthand for making sense of a result: knowing the mean salary and standard deviation lets you place where a specific salary sits in the distribution.
For skewed data (incomes, home prices, response times) the standard deviation is still mathematically defined, but the empirical rule does not apply and the mean itself is pulled by extreme values. In those cases, pair the standard deviation with the median and interquartile range for a fuller picture.
Worked example
2, 4, 4, 4, 5, 5, 7, 9
- Mean: (2 + 4 + 4 + 4 + 5 + 5 + 7 + 9) ÷ 8 = 40 ÷ 8 = 5.
- Squared deviations from the mean: (2−5)² + (4−5)² × 3 + (5−5)² × 2 + (7−5)² + (9−5)² = 9 + 3 + 0 + 4 + 16 = 32.
- Population variance: 32 ÷ 8 = 4. Population standard deviation σ = √4 = 2.
- Sample variance: 32 ÷ (8 − 1) = 32 ÷ 7 = 4.571. Sample standard deviation s = √4.571 ≈ 2.138.
Common mistakes
- Using the population formula on sample data. The underestimation is small for large n but material for n under 30, and most research datasets are samples.
- Mixing up variance and standard deviation. Variance is in squared units (dollars-squared, metres-squared) and is not directly interpretable; the standard deviation is in the original units.
- Applying the 68-95-99.7 rule to clearly skewed data. For incomes or home prices, two standard deviations above the mean is not where "97.5% of values sit below" — the distribution is one-sided.
Limitations of this calculator
- Calculates on numeric data entered as a flat list. Grouped data, frequency tables and weighted datasets need their own formulas.
- Does not detect outliers automatically. A single typo of 7000 instead of 7.000 can dominate the result.
- Reports to three decimal places; downstream statistical tests may need more precision.
For your specific situation, consult a tax professional or financial advisor.
Questions people ask
Why divide by n − 1 for a sample?
A sample's values sit closer to their own mean than to the true population mean (because the sample mean is calculated from exactly those values). Dividing by n would therefore underestimate the real population spread. Bessel's correction, dividing by n − 1 instead, removes that bias on average across many samples.
What does the standard deviation actually mean in practice?
Think of it as a typical distance from the mean. If a class's mean test score is 72 with a standard deviation of 8, a score of 80 is "one standard deviation above the mean" — above average but inside the normal range of variation. For roughly normal data, about 68% of scores fall within one standard deviation and 95% within two.
What is variance and why do we square the deviations?
Variance is the standard deviation squared. We square the deviations in the formula because unsquared deviations average to zero (positives and negatives cancel). Squaring also weighs big deviations more heavily, which captures how much a single far-out value spreads the data. The square root at the end puts the answer back in the original units.
Standard deviation or interquartile range?
Standard deviation is best for roughly symmetric data. For skewed data (incomes, waiting times) or data with extreme outliers, use the median and interquartile range (IQR) instead. Both quantities ignore the extremes and describe the middle 50% of the data.
Sources
Last reviewed October 2, 2026 · Last reviewed by Muhammad Ahmad on October 31, 2026