Standard deviation (SD) and standard error (SE) are two of the most confused quantities in statistics — partly because SE is calculated from SD, and partly because reports often use them interchangeably. They measure different things, and picking the wrong one can badly mislead a reader.
Standard deviation: how spread out the data is
The standard deviation describes the variability of the individual values in your dataset — how far a typical data point sits from the mean. It's a property of the data itself and does not shrink as you collect more of it. If adult heights have an SD of 7 cm, that stays about 7 cm whether you measure 30 people or 30,000.
Standard error: how precise your estimate of the mean is
The standard error (of the mean) describes the uncertainty in your estimate of the average, not the spread of the data. It answers: "if I repeated this study, how much would the sample mean bounce around?" Crucially, it shrinks as your sample grows:
SE = SD / √n
Because of the √n in the denominator, quadrupling your sample size halves the standard error. More data doesn't make your subjects less variable (SD stays put) — it makes your estimate of their average more precise (SE drops).
Side by side
| Standard deviation | Standard error | |
|---|---|---|
| Describes | Spread of individual values | Precision of the mean estimate |
| Changes with n? | No (stable) | Yes — shrinks as n grows |
| Use it to… | Describe your data / variability | Build a confidence interval or run a test |
Which should you report?
- Describing your sample (how variable are the measurements?) → report the SD.
- Making an inference about the population mean (how confident are you in the average?) → use the SE, usually as a confidence interval.
A common bad habit is putting ± SE error bars on a chart to make results look more precise than they are. Always label which one you're showing.
Worked example. Test scores: mean 70, SD 10, n = 100. The SD (10) says individual scores typically fall within ±10 of 70. The SE = 10/√100 = 1.0, so a 95% CI for the true mean is roughly 70 ± 1.96×1.0 = [68.0, 72.0]. The data is spread over tens of points, but the average is pinned down to within a point.
Calculate both
Descriptive Statistics Calculator →Reports SD and the standard error of the mean together from your data.Related reading: Confidence intervals explained · Mean, median & mode · What is statistical significance?
Sources & further reading: NIST/SEMATECH e-Handbook of Statistical Methods · OpenStax Introductory Statistics. How we compute & validate: methodology. Last reviewed August 2026.