Standard deviation summarizes how widely values vary around their mean. It is not automatically good or bad; its meaning depends on the units, population, and question being asked.
How the measure is built
The calculation finds each value's distance from the mean, squares those distances, averages them under the selected population or sample convention, and takes the square root. The final unit matches the original data.
Population versus sample
A population standard deviation describes the complete group being studied. A sample standard deviation uses a degrees-of-freedom adjustment when the data are treated as a sample of a larger group.
Do not use it without distribution context
Rules such as 68% within one standard deviation assume a roughly bell-shaped distribution. Skewed data, outliers, and small samples need more careful interpretation.
A worked example
Worked example: two data sets can have the same mean but different standard deviations if one is tightly grouped and the other has observations spread farther away. The unit remains the unit of the original measurements.
Plot or inspect the observations when possible. One large outlier can change the standard deviation substantially without explaining why the data vary.
Spread should be read alongside sample size and the shape of the data. A standard deviation alone cannot tell you whether a process is stable, biased, or improving.
Standard deviation is a compact spread measure, not a complete description of a distribution or its causes.
COMMON QUESTIONS
Frequently asked questions
Can standard deviation be negative?
No. It is based on squared deviations and is zero only when all values are identical.
Why do sample and population results differ?
They use different denominators because they answer different questions about complete data versus an observed sample.