Skip to content
M2TOOLKIT

Calculators · free tool

Sample Size Calculator

Work out how many responses you need before you run a survey or study.

Use it as often as you like — it's free, with no sign-up and no limits.

  • Calculators
  • Free
  • No sign-up
  • Nothing is sent or stored
  • Updated
You're estimating a
%
%

Use 50% if you don't know — it gives the safest (largest) sample.

For small populations the required sample is smaller.

Result

Sample size needed
385
Before population correction
385
Critical z
1.96
How it was calculated
  1. n₀ = z² × p(1 − p) ÷ E² = 1.96² × 0.5 × 0.5 ÷ 0.05²
  2. = 384.1459
  3. Round up → 385

The calculation happens instantly in your browser. The numbers you enter are not sent to our servers or saved.

How to use the Sample Size Calculator

  1. Choose whether you're estimating a percentage or an average.
  2. Set the confidence level and margin of error.
  3. Optionally enter the population size for small populations.

What does this tool do?

To estimate a percentage within ±5% at 95% confidence you need about 385 responses, no matter how big the population — as long as it's large. Halving the margin of error roughly quadruples the sample you need.

If your whole population is small (say 500 employees), the finite population correction reduces the required sample. Using 50% as the expected proportion is the safe choice when you don't know it.

Worked examples from the calculator

A survey percentage at 95% confidence, ±5% margin, unknown proportion (use 50%): n = 1.96² × 0.5 × 0.5 ÷ 0.05² = 384.15, rounded up to 385. Tighten the margin to ±2.5% and you need 1,537, about four times as many, because halving the margin quadruples the sample.

If your whole population is small, say 500 employees, the finite population correction n₀ ÷ (1 + (n₀ − 1) ÷ N) brings 385 down to 218. For an average rather than a percentage, you need an estimate of the standard deviation: with σ = 15 and a margin of ±3 units at 95%, n = (1.96 × 15 ÷ 3)² = 96.04, so 97.

Responses needed vs people to invite

The result is the number of completed responses you need, not the number of people to contact. If you expect 20% of people to respond, invite 385 ÷ 0.2 = 1,925. Response rates for email surveys are often well below that, so plan from your own past rates where you have them.

Related: Confidence Interval Calculator

Why use it?

  • Survey and measurement modes.
  • Finite population correction.
  • Shows the calculation step by step.

Use cases

  • Customer and employee surveys.
  • Student research projects and dissertations.
  • Market research and polling.
  • Planning A/B tests for a conversion rate (as a rough first estimate).

Common mistakes

  • Guessing an expected proportion near 0% or 100% without evidence; 50% is the safe default because it gives the largest sample.
  • Reporting results for subgroups (for example, one department) with the margin of error calculated for the whole sample.

The formula

Proportion: n₀ = z² p(1 − p) ÷ E²
Mean: n₀ = (z σ ÷ E)²
Finite population: n = n₀ ÷ (1 + (n₀ − 1) ÷ N)

Accuracy and limits

  • The formulas assume simple random sampling. Stratified or clustered designs need a design-effect adjustment that this calculator doesn't apply.

Privacy

The calculation happens instantly in your browser. The numbers you enter are not sent to our servers or saved. There's no account to create and nothing to install.

Frequently asked questions

Does a bigger population need a bigger sample?

Barely. Beyond a few tens of thousands, the required sample hardly changes.

Content last updated . See how M2Toolkit works.

Other tools people use alongside the sample size calculator.