Free tools that run in your browser. Use any tool without an account.

Favourites Cart
Sign in
Join free

Sample Size Calculator

  • Free
  • Runs in your browser
  • No account needed

For a 95% confidence level and ±5% margin of error you need about 385 completed responses using p = 0.5; the tool shows the required sample size, the exact pre-correction value, z used, and FPC-adjusted n.

Sample size needed
385
Before the population correction
385
Exact value before rounding up
384.15
z value used
1.96

This is the number of completed responses you need. If you expect only some people to reply, divide by the response rate (for example 385 / 0.4 = 963 invitations at a 40% response rate).

Runs in your browser. Your files and values are not sent to us. Report a problem
Advertisement

About the Sample Size Calculator

This free calculator estimates how many completed responses you need to measure a percentage or an average. You pick whether you are estimating a percentage (proportion) or an average (mean). Enter confidence level, margin of error, expected percentage or expected standard deviation, and population size if known. Results update instantly in your browser and nothing is uploaded. The page shows the sample size rounded up, the exact pre-rounding value, the value before population correction, and the z critical value used. It uses standard Cochran and mean formulas with finite population correction. You can copy results, and share a link that reopens with your numbers.

How to use it

  1. 1Choose whether you are estimating A percentage (survey, poll) or An average (mean).
  2. 2Enter confidence level (%) and margin of error (%) or margin in your units.
  3. 3Type expected percentage or expected standard deviation and population size if known.
  4. 4Read the sample size needed, the exact pre-round value, the before-correction value, and the z value used.
Advertisement

Worked example

Estimating a percentage: Confidence level 95%, Margin of error 5%, Expected percentage 50%, Population size 10000. Before population correction n0 = 384.16. Exact value before rounding = 370.04. z used = 1.96. Sample size needed = 371 completed responses.

The formula

Percentage: n0 = z^2 x p(1 - p) / E^2, with p and E as decimals (50% = 0.5)Mean: n0 = (z x SD / E)^2With a known population N, the finite population correction gives n = n0 / (1 + (n0 - 1) / N)Always round upz is 1.96 for 95% confidence, 2.576 for 99%

Source: Cochran, Sampling Techniques (formula background)NIST, Handbook of Statistical Methods (z values for CI)Pearson sample size calculator (method and FPC explained). Last checked 6 November 2026.

What it is good for

  • Shows rounded required completed responses and the exact pre-rounding value.
  • Applies finite population correction when you enter population size.
  • Runs entirely in your browser; nothing is uploaded.
  • Does not perform power analysis or A/B test sample sizes.
  • Does not adjust automatically for cluster sampling or design effects.
Advertisement

Questions people ask

How many responses do I need for 95% confidence and ±5% margin of error?

About 385 completed responses using p = 0.5. That comes from n0 = 1.96^2 x 0.5(1-0.5) / 0.05^2 = 384.16, rounded up to 385.

When should I enter a population size?

Enter population size when your sample will be a non-negligible share of the population. Use it if sample/N exceeds about 0.05 to apply the finite population correction.

How do I pick the expected percentage p if I do not know it?

Use 50% (p = 0.5) for a conservative estimate. That maximizes p(1-p) and gives the largest required sample size.

Do I need to invite more people because some will not reply?

Yes. This calculator gives the number of completed responses needed. Divide by expected response rate to get invitations. Example: 385 / 0.4 = 963 invites at 40% response.

Can I use this for A/B tests or power calculations?

No. This tool estimates precision for a single proportion or mean. Use a power or two-sample calculator for hypothesis tests and A/B experiments.

Where do my inputs and results go?

All calculations run in your browser and do not upload your data. You can copy results or save a link that reopens with your numbers.

Help keep every tool free

FreeToolr stays free because people help it grow. Share a tool you like, tell us what to build next, or support us on Ko-fi.

Advertisement
Advertisement