Math / calculator
Standard Deviation Calculator
Calculate count, sum, mean, population and sample variance and standard deviation from numeric values. Stable arithmetic distinguishes n from n−1 denominators.
Calculator
Calculator
Calculate population and sample dispersion.
Enter decimal or scientific-notation values separated by commas, spaces or newlines, without thousands separators. One value has zero population dispersion; sample dispersion requires at least two.
Your result
Enter details to see the result.
Complete the inputs and calculate to see the main result and its supporting breakdown.
Example
Worked example
Example inputs
- Numeric values
- 2, 4, 4, 4, 5, 5, 7, 9
Calculation breakdown
Mean=Σx/n. M2=Σ(x−mean)²; population variance=M2/n and sample variance=M2/(n−1). Standard deviation is the square root of the corresponding variance. An origin-shifted Welford update avoids subtracting nearly equal large sums of squares.
- Count
- 8
- Sum
- 40
- Mean
- 5
- Population variance (divide by n)
- 4
- Population standard deviation
- 2
- Sample variance (divide by n−1)
- 4.571429
Example result
Sample standard deviation
2.13809
How to interpret the result
Population and sample outputs use different denominators. Standard deviation retains the data unit and variance uses squared units; the one-value sample message is intentional because n−1 is zero, rather than a numeric dispersion result.
How to use this calculator
- Paste 1–1000 decimal or scientific-notation numbers separated by commas, spaces or newlines.
- Use a decimal point with no thousands separators, and include only observations you intend to analyze.
- Calculate and choose the population or sample interpretation according to what the dataset represents.
How it works
Formula
Mean=Σx/n. M2=Σ(x−mean)²; population variance=M2/n and sample variance=M2/(n−1). Standard deviation is the square root of the corresponding variance. An origin-shifted Welford update avoids subtracting nearly equal large sums of squares.
Should I use sample or population standard deviation?
Use the population denominator n when the values are the complete population being described. Use the sample denominator n−1 for the usual sample variance estimate. Standard deviation restores the original units, while variance squares them; one observation cannot define sample variance.
What this calculator does
Standard deviation measures dispersion around the mean in the original data unit; variance has squared units. Whether values represent the whole population or a sample changes the denominator. Both interpretations are reported without assuming which dataset role you intended.
Who this is useful for
Students and analysts summarizing the spread of a finite numeric list while explicitly distinguishing a complete population from a sample.
Inputs you will need
Required or selected inputs
- Numeric values
Assumptions
- Values are equally weighted numeric observations, with commas or whitespace as separators and decimal points rather than thousands separators.
- The population calculation uses every entered value; sample dispersion requires at least two observations.
Limitations
- No missing-value imputation, confidence interval, outlier removal or normal-distribution assumption is added.
- Input values are finite, bounded and limited to 1000 entries. Floating-point arithmetic cannot recover differences already lost when numbers were entered or rounded.
Useful next steps
Add median, minimum and maximum for a complementary center-and-range summary with Average calculator.
Evaluate a separate formula using the reported mean or dispersion with Scientific Calculator.
References
NIST: measures of scale — Reference definitions for variance and standard deviation. Sample variance uses n−1; population variance uses n.
See how HandyChest handles units, validation, rounding, and testing, or Report a problem with this calculator.
Good to know
Frequently asked questions
Why is stable variance arithmetic useful?
The naive difference between a sum of squares and a squared sum can erase small spread when values have a large common offset. The shifted incremental algorithm works with differences and avoids that cancellation pattern.
Keep going