Enter a comma- or space-separated list of numbers to compute the sample standard deviation s with n - 1 denominator, assessing how spread out the observations are.
输入多个数值（逗号或空格分隔），求样本标准差。
| Item | Formula / Note | Source / Standard |
|---|---|---|
| Sample standard deviation | s = sqrt( sum(x_i - xbar)^2 / (n - 1) ) | ISO 3534-1:2006, Statistics - Vocabulary and symbols |
| Bessel correction | Divide by n - 1 rather than n to obtain an unbiased variance estimate | ISO 3534-1:2006; Cramer H., Mathematical Methods of Statistics (1946) |
| Empirical rule | For a normal population roughly 68%/95%/99.7% lie within ±1/±2/±3 sigma | NIST/SEMATECH e-Handbook of Statistical Methods, normal distribution |
| Chebyshev inequality | For any distribution at least 1 - 1/k^2 of data lie within k sigma of the mean (k > 1) | Chebyshev P. (1867), Des valeurs moyennes |