Standard Deviation Calculator
Compute sample (s) or population (σ) standard deviation, variance, mean, median, standard error, and step-by-step squared deviations.
Standard Deviation Calculator
Change the inputs that matter to your decision. Optional assumptions stay collapsed until you need them.
Dataset
Enter numbers separated by commas, spaces, semicolons, or new lines.
Mean 21.4286, variance (s²) 45.2857, and range 18 across 7 observations.
| Count (n) | 7 |
|---|---|
| Sum (Σx) | 150 |
| Minimum / Maximum | 12 / 30 |
| Sum of squared deviations | 271.7143 |
| Coefficient of variation | 31.4042% |
| x1 = 12 | (x − mean)² = 88.898Deviation: -9.4286 |
| x2 = 15 | (x − mean)² = 41.3265Deviation: -6.4286 |
| x3 = 18 | (x − mean)² = 11.7551Deviation: -3.4286 |
| x4 = 22 | (x − mean)² = 0.3265Deviation: +0.5714 |
| x5 = 25 | (x − mean)² = 12.7551Deviation: +3.5714 |
| x6 = 28 | (x − mean)² = 43.1837Deviation: +6.5714 |
| x7 = 30 | (x − mean)² = 73.4694Deviation: +8.5714 |
Formula: s = √( Σ(x − x̄)² ÷ (n − 1) ).
Sample (n − 1) vs. population (n)
Why Bessel's correction divides by n − 1 when estimating spread from a sample.
When your numbers are a sample drawn from a larger population, the sample mean is closer to those specific points than the true population mean is, which slightly understates squared deviations. Dividing by n − 1(Bessel’s correction) yields an unbiased variance estimator. When your list contains every value in the population, divide by n.
Sample datasets to compare
Pre-filled datasets illustrating low vs. high dispersion.
Frequently asked questions
Practical notes about the inputs, assumptions, and result.
Choose Sample when your numbers are a subset used to estimate a broader group (the usual choice for experiments, surveys, and lab data). Choose Population only when you have measured every single member of the entire group.
Because the sample mean is calculated from the same data points, deviations around it are slightly smaller on average than deviations around the true population mean. Dividing by n − 1 (Bessel's correction) removes that downward bias from variance.
Variance is the average squared deviation (in squared units, such as dollars² or cm²). Standard deviation is the square root of variance, returning spread to the same units as your original data.
Standard deviation measures how widely individual data points scatter around the mean. Standard error (SD ÷ √n) measures how precisely your sample mean estimates the true population mean.
It means every value in the dataset is identical, so there is zero variation around the mean.
CV is standard deviation divided by the absolute mean (expressed as a percentage), letting you compare relative variability across datasets with different scales or units.
Glossary
Definitions for the conventions and terms used above.
- Bessel's correction (n − 1)
- Using n − 1 degrees of freedom in sample variance so the estimate does not systematically understate population variance.
- Variance (s² or σ²)
- The average of the squared deviations from the mean, before taking the square root.
- Standard error (SE)
- Standard deviation divided by √n, indicating the expected spread of the sample mean across repeated samples.