Standard Deviation of Sample Mean Calculator

Standard Deviation of Sample Mean

Calculate the standard error of the sample mean with optional finite population correction for small populations.

Last updated: March 2026

Calculator

Known population variability

Number of observations

Leave blank for infinite population

Standard Error (σ_x̄)
2.738613
Without population adjustment
Adjusted SE (with FPC)
2.657843
Finite Population Correction applied
Formula (infinite)σ / √n
Population SD (σ)15.0000
Sample Size (n)30
√n5.4772
Population Size (N)500
FPC√((N-n)/(N-1)) = 0.970507
Formula (finite)σ / √n × FPC

What is Standard Deviation of Sample Mean?

The Standard Deviation of the Sample Mean (σ_x̄), also called Standard Error of the Mean, measures how much sample means vary from the true population mean. When you repeatedly draw samples from a population, each sample has a different mean. This metric quantifies that variation.

The formula is: σ_x̄ = σ / √n, where σ is the population SD and n is the sample size. For large populations, this formula works well. However, for small populations (finite populations), we apply a Finite Population Correction (FPC): σ_x̄ = (σ / √n) × √((N-n)/(N-1)). This correction reduces the standard error because we're sampling without replacement from a limited population, making the sample more representative.

The FPC matters when the sample size is a significant fraction of the population (typically when n > 0.05N). For large populations, FPC ≈ 1 and the correction becomes negligible.

How to Calculate Standard Deviation of Sample Mean

For Large/Infinite Populations

Step 1: Identify population SD (σ) and sample size (n)
Step 2: Divide σ by the square root of n: σ_x̄ = σ / √n

For Finite Populations (Small)

Step 1: Calculate basic SE: σ_x̄ = σ / √n
Step 2: Calculate FPC: FPC = √[(N − n) / (N − 1)]
Step 3: Apply correction: σ_x̄ = (σ / √n) × FPC

Formulas

Infinite Population:
σ_x̄ = σ / √n
Finite Population:
σ_x̄ = (σ / √n) × √[(N − n) / (N − 1)]
When to Use FPC:
Apply FPC when n > 0.05N (sample is > 5% of population)

Real-World Example

Surveying a Small School District

Context:
A district superintendent knows that standardized test scores across all schools have σ = 8 points. The district has only 120 schools total. To estimate the average test score, the superintendent randomly samples 20 schools.
Calculation:
WITHOUT FPC (treating as infinite):
σ_x̄ = 8 / √20 = 8 / 4.47 = 1.79 points
WITH FPC (finite population correction):
n = 20, N = 120, n/N = 20/120 = 16.7% > 5%
FPC = √[(120 − 20) / (120 − 1)] = √[100/119] = 0.917
σ_x̄ = 1.79 × 0.917 = 1.64 points
The corrected estimate (1.64) is < the unadjusted (1.79) because sampling 20 out of 120 schools without replacement ensures better representation. The 0.15-point difference might seem small but is meaningful in decision-making.

Frequently Asked Questions

What's the difference between σ and σ_x̄?

σ (population SD) measures how spread out individual values are. σ_x̄ (SE of mean) measures how spread out sample means are. σ_x̄ is always smaller than σ by a factor of √n — sample means are more stable than individual values.

When do I need to use FPC?

Use FPC when sampling without replacement from a small/finite population. A practical rule: apply FPC if n {'>'} 0.05N (sample {'>'} 5% of population). For large populations or sampling with replacement, FPC ≈ 1 and can be ignored.

Why does larger sample size reduce standard error?

Larger samples provide better estimates of the population mean. Mathematically, σ_x̄ includes √n in the denominator, so as n increases, SE decreases. Doubling n reduces SE by about 30%; quadrupling n cuts SE in half.

What does it mean if FPC = 0.95?

FPC = 0.95 means the corrected SE is 95% of the uncorrected SE. You're saving only 5% of error by the population adjustment, indicating the finite population effect is small. This occurs when n is a small fraction of N.

Can standard error of mean ever be zero?

Theoretically, only if σ = 0 (no variation in population) or n = ∞ (infinite sample). In practice, SE is never zero because real populations always have variation, and infinite samples are impossible.

How does this relate to confidence intervals?

A 95% CI for the population mean = sample mean ± 1.96 × σ_x̄ . The standard error directly scales the width of confidence intervals. Smaller SE means narrower, more precise confidence intervals.

Related Tools