CHART CHOICES & CHECKABLE METHODS

Standard deviation vs standard error: spread and precision

Standard deviation describes variation among observations. Standard error describes the sampling variability of an estimator. For an independent identically distributed sample with finite variance, the estimated standard error of the mean is sample SD divided by √n.

All chart guides →

The comparison at a glance

QuantityDescribesExample
Sample SDObservation spread2.582
SE of meanEstimated mean sampling variability1.291
IQRMiddle-half spreadDifferent statistic
Confidence intervalMethod-specific uncertainty intervalNot calculated here

Work a four-value example

For fictional observations 2, 4, 6 and 8, the mean is 5. Deviations are −3, −1, 1 and 3; squared deviations sum to 20. The sample variance uses n − 1 = 3, giving 20/3. Sample SD is √(20/3), approximately 2.582. The estimated standard error of the mean is 2.582/√4, approximately 1.291.

Keep the denominator convention visible

Dividing the squared-deviation sum by four would instead give a population-style variance of 5 and SD √5. That is a different convention. Here the calculation uses sample SD because it feeds an estimated standard error. State which formula you used when reproducing a result in a spreadsheet or comparing published summaries.

More rows do not always mean more information

The s/√n formula assumes an appropriate sampling structure. Repeated records for the same individual, clustered observations or serial dependence can invalidate the naive treatment of every row as independent. Copying the same four values multiple times into a spreadsheet does not create new independent evidence or justify a more confident conclusion.

Choose and label error bars carefully

An SD bar and an SE bar communicate different quantities, and neither is automatically a 95% confidence interval. A confidence interval needs a stated method and assumptions. ChartsAI’s current box plots show quartiles and whiskers rather than SD, SE or confidence intervals; do not relabel those summaries as uncertainty bars.

A related chart you can inspect

Side-by-side box plots comparing two groups
Side-by-side box plots comparing two groups. Original fictional data; this linked example has its own complete input and calculation settings.

Open the worked example ↗

References and reproducibility

The worked arithmetic and teaching examples on this page are original. Background reading: NIST measures of scale. Source code and reuse terms expose the implemented calculations.

Written by SupaMakers for ChartsAI · September 8, 2026.

Continue with a related question