What is the difference between standard deviation and standard error?

Standard deviation (SD) measures how spread out individual data points are within a single sample or population. Standard error (SE) measures how spread out the sample statistic (like the mean) would be across many repeated samples. SD describes variability in data; SE describes variability in an estimate. Numerically, SE = SD / √n, so SE is always smaller than SD when n > 1.

Standard Error — Definition, Formula & Examples

Standard error is a measure of how much a sample statistic (like the sample mean) would vary if you repeatedly drew new random samples from the same population. A smaller standard error means your statistic is a more precise estimate of the true population parameter.

The standard error of a statistic is the standard deviation of its sampling distribution. For the sample mean $\bar{x}$ , the standard error equals the population standard deviation $\sigma$ divided by the square root of the sample size $n$ . When $\sigma$ is unknown, it is replaced by the sample standard deviation $s$ , yielding an estimated standard error.

Key Formula

SE = \frac{s}{\sqrt{n}}

Where:

$SE$ = Standard error of the sample mean
$s$ = Sample standard deviation (used when the population standard deviation σ is unknown)
$n$ = Sample size (number of observations)

How It Works

Standard error tells you how "noisy" your estimate is. After collecting a sample, you compute a statistic such as the sample mean. Because every sample is different, that statistic has its own distribution — the sampling distribution — and the standard error is the spread of that distribution. To calculate it, divide the standard deviation by the square root of your sample size. Larger samples produce smaller standard errors, which means more precise estimates. You use SE directly when building confidence intervals (statistic

\pm

critical value

\times

SE) and when computing test statistics for hypothesis tests.

Worked Example

Problem: A random sample of 36 students has a mean test score of 78 with a sample standard deviation of 12. Find the standard error of the mean.

Identify the values: The sample standard deviation is 12 and the sample size is 36.

s = 12, \quad n = 36

Apply the formula: Divide the sample standard deviation by the square root of the sample size.

SE = \frac{12}{\sqrt{36}} = \frac{12}{6}

Compute: The standard error equals 2.

SE = 2

Answer: The standard error of the mean is 2 points. This means that if you repeatedly sampled 36 students, the sample means would typically differ from the true population mean by about 2 points.

Another Example

Problem: Suppose you increase the sample size to 144 students, keeping the same standard deviation of 12. What happens to the standard error?

Apply the formula with the new sample size: Use the same standard deviation but a larger n.

SE = \frac{12}{\sqrt{144}} = \frac{12}{12} = 1

Compare: Quadrupling the sample size (from 36 to 144) cut the standard error in half (from 2 to 1).

Answer: The new standard error is 1. This illustrates that to halve the SE, you must quadruple the sample size, because SE decreases with

\sqrt{n}

, not

n

Visualization

Why It Matters

Standard error is central to AP Statistics and any college-level course that involves inference. Every confidence interval you build and every hypothesis test you run uses SE to convert a raw difference into a standardized test statistic. In fields like clinical research and polling, SE determines whether a result is precise enough to guide real decisions — a poll's reported margin of error, for instance, is directly based on the standard error of the sample proportion.

Common Mistakes

Mistake: Confusing standard deviation with standard error and reporting SD where SE is needed.

Correction: Standard deviation describes spread within your data set. Standard error describes precision of a statistic. When constructing confidence intervals or test statistics, always divide by √n to get SE.

Mistake: Believing that doubling the sample size cuts the standard error in half.

Correction: SE decreases by the square root of n. To halve SE, you must multiply the sample size by 4, not 2. For example, going from n = 25 to n = 100 halves the SE.