What is standard error?

The standard error of the mean (SEM) measures how much the sample mean would vary if you repeated your study with different random samples from the same population. SEM = s / √n, where s is the sample standard deviation and n is the sample size. Unlike standard deviation (which describes the spread of individual data points), standard error describes the precision of the sample mean as an estimate of the true population mean. A smaller SEM means your estimate of the mean is more precise.

How is standard error different from standard deviation?

Standard deviation (s) measures how spread out individual data points are from the sample mean. Standard error (SEM = s / √n) measures how much the sample mean itself would vary across repeated samples. Standard deviation is a property of your data; standard error is a property of your estimate. For example, with s = 10 and n = 100: SEM = 10 / √100 = 1.0. The SD tells you individual values typically differ from the mean by 10 units; the SEM tells you the sample mean is typically within 1 unit of the true population mean.

How do you calculate standard error?

SEM = s / √n, where s is the sample standard deviation and n is the sample size. This calculator offers two modes: (1) Enter s and n directly if you already have summary statistics; (2) Enter raw data and the calculator computes s first, then SEM. For raw data {12, 15, 14, 18, 11, 16}: mean = 14.33, s = 2.58, n = 6, SEM = 2.58 / √6 = 1.05. This means the sample mean of 14.33 is estimated to be within about ±1 unit of the true population mean.

What does a smaller standard error mean?

A smaller SEM means the sample mean is a more precise estimate of the true population mean. It indicates that if you repeated the study with a new random sample, the new mean would likely be close to the current one. You can achieve a smaller SEM by increasing the sample size n (since SEM ∝ 1/√n), by reducing within-group variability (by controlling experimental conditions), or both. In medical research, a smaller SEM allows detection of smaller treatment effects with the same confidence.

How does sample size affect standard error?

SEM decreases as sample size increases, but with diminishing returns. Specifically, SEM = s / √n, so doubling n reduces SEM by a factor of √2 ≈ 1.41 (not by half). To halve the SEM, you must quadruple n. For example, with s = 10: n = 25 gives SEM = 2.0; n = 100 gives SEM = 1.0; n = 400 gives SEM = 0.5. This diminishing return explains why large studies are expensive — each additional reduction in SEM requires an exponentially larger sample.

What is standard error used for?

Standard error is the foundation of confidence intervals and hypothesis tests. The 95% confidence interval for a mean is x̄ ± 1.96 × SEM. It is used in reporting scientific results — error bars on graphs typically show ±1 SEM or ±2 SEM. In medicine, SEM appears in clinical trial reports to show how precisely the treatment effect was measured. In polling, the "margin of error" is essentially a confidence interval built from the standard error of a proportion. The smaller the SEM, the more conclusive the study.

Why is standard error smaller than standard deviation?

SEM is always smaller than SD (for n > 1) because SEM = SD / √n. This makes sense intuitively: individual data points are much more variable than sample averages. If you measured the height of 1 person, you might get anything from 4'7" to 6'8" (high SD). But if you averaged 100 people's heights, your average would almost certainly land between 5'5" and 5'9" (low SEM). Averages cancel out extreme values, so the distribution of means is always narrower than the distribution of individual values.

When should I report SD vs. SEM in a paper or report?

Use standard deviation (SD) when describing the characteristics of your sample or population — for example, "participants had a mean age of 34.2 ± 8.7 years (SD)." Use SEM when describing the precision of a mean estimate — for example, "the treatment reduced blood pressure by 6.4 ± 1.2 mmHg (SEM)." Many researchers use SEM because it looks smaller and makes results appear more precise, but this is misleading if your goal is to characterize variability in the population. When in doubt, report SD and let readers compute SEM themselves.

Standard Error Calculator — SEM from Data or SD

Calculates the standard error of the mean (SEM) from raw data or from SD and sample size — shows 95% confidence interval.

How to Use the Standard Error Calculator

This standard error calculator finds the SEM and 95% confidence interval from either raw data or summary statistics — choose between two input modes: From SD & n — enter the sample standard deviation and sample size directly — or From raw data — enter a comma-separated list of values and the calculator computes the sample SD and SEM automatically. In raw data mode, the calculator also displays the sample mean and 95% confidence interval (x̄ ± 1.96 × SEM).

Standard error builds directly on standard deviation. If you need the sample standard deviation first, use our standard deviation calculator, then paste s and n into this calculator's summary mode. To build a full confidence interval from the SEM, see our confidence interval calculator.

The Standard Error Formula

The standard error of the mean is:

SEM = s / √n

Where s is the sample standard deviation and n is the sample size. The formula shows that SEM decreases as n increases — larger samples produce more precise mean estimates.

Worked Example

A nutritionist measures the daily protein intake (grams) for 9 participants: 72, 68, 85, 74, 66, 78, 80, 71, 76.

Mean (x̄) = (72 + 68 + 85 + 74 + 66 + 78 + 80 + 71 + 76) / 9 = 670 / 9 ≈ 74.44 g
Sample SD (s) ≈ 6.28 g
SEM = 6.28 / √9 = 6.28 / 3 ≈ 2.09 g
95% CI: 74.44 ± 1.96 × 2.09 = [70.34, 78.54] g

Interpretation: We are 95% confident the true mean daily protein intake for this population falls between 70.34 g and 78.54 g.

AdvertisementResponsive Ad

Standard Error vs. Standard Deviation

These two statistics are closely related but measure different things, and confusing them is one of the most common errors in statistics reporting.

Standard deviation (s):Describes the spread of individual data points around the sample mean. It answers: "How variable is the data?" SD does not shrink as n increases — adding more observations doesn't reduce the natural variability in the population.
Standard error (SEM = s/√n):Describes the precision of the sample mean as an estimate of the population mean. It answers: "How precisely do I know the mean?" SEM always shrinks as n increases — more data gives a more precise mean estimate.

In scientific papers, results are often reported as "mean ± SD" (describing the data spread) or "mean ± SEM" (describing the precision of the mean estimate). The choice matters: SEM is always smaller than SD for n > 1, so reporting ± SEM makes results appear more precise. Use ± SD when describing the distribution of a population; use ± SEM when describing the uncertainty in a mean estimate.

Sample Size and Precision

Because SEM = s / √n, the relationship between sample size and precision is non-linear:

Doubling n reduces SEM by a factor of √2 ≈ 1.41 (a 29% reduction)
Quadrupling n halves the SEM
A 100× increase in n reduces SEM by a factor of 10

This means there is a law of diminishing returns for precision. Going from n = 10 to n = 40 halves the SEM, but going from n = 40 to n = 90 only reduces it by another 33%. Researchers must balance precision against the cost of collecting more data.

A common rule of thumb in clinical research is n ≥ 30 per group, which provides enough statistical power for many parametric tests and ensures SEM is a reliable precision estimate.

AdvertisementResponsive Ad

Standard Error in Error Bars

In scientific graphs, error bars show the variability or uncertainty around data points. Error bars can represent ± standard deviation, ± standard error, or ± confidence interval — and the choice dramatically affects how the graph looks and what it communicates.

± SD error bars: Show the spread of the data. They describe the population distribution and do not shrink with more samples.
± SEM error bars: Show the precision of the mean estimate. They get smaller as n increases. Often used in biology and medicine.
± 95% CI error bars: Show the confidence interval for the mean. If two 95% CI bars do not overlap, the difference between groups is statistically significant at p < 0.05 (approximately).

When reading a graph with error bars, always check the figure legend to determine which type of error bar is shown. SEM bars look narrower than SD bars, which can make results appear more consistent than they are. The American Psychological Association and many journals now require authors to specify which error bar type is used.

Sources & References

Standard Error — NIST/SEMATECH e-Handbook of Statistical Methods
Standard Error of the Mean — Khan Academy — Khan Academy

Standard Error Calculator