Sample Size for Proportion Calculator

How Many Responses Do You Actually Need?

Every survey, poll, or A/B test estimating a proportion — a conversion rate, a yes/no vote share, a defect rate — has to answer this question before data collection starts: how many observations are enough to trust the result? Collect too few and the margin of error swamps any real signal; collect far more than necessary and the extra effort is wasted. This calculator works out the minimum sample size for a target margin of error and confidence level.

The Formula

n = (Z² × p × (1−p)) / E²
where Z is the critical value for the chosen confidence level, p is the expected proportion, E is the margin of error (as a decimal)

Sample Size at Various Margins of Error

Assuming the most conservative case, p = 0.5 (maximum variance), and 95% confidence:

Required sample size by margin of error, p = 0.5, 95% confidence (z = 1.96)
Margin of ErrorRequired n
±10%97
±5%385
±3%1,068
±2%2,401
±1%9,604

Sample sizes computed directly from the calculator's formula and rounded up to the next whole respondent.

Halving the margin of error roughly quadruples the required sample — a direct consequence of E appearing squared in the denominator, and the reason tightening a poll from ±3% to ±1% is so much more expensive than it sounds.

Sample Size by Confidence Level

Required sample size at a fixed ±5% margin, p = 0.5
Confidence LevelZRequired n
90%1.645271
95%1.96385
99%2.576664

Where This Calculation Matters

  • Political and market polling — the familiar "margin of error of plus or minus 3 points" quoted with a poll traces directly back to this formula.
  • Conversion rate testing — before launching an A/B test on a website, this determines how many visitors are needed per variant to detect a meaningful difference in conversion rate.
  • Quality control sampling — estimating a defect rate in a manufacturing batch to a required precision uses the same formula.

How to Use This Calculator

  1. Select a confidence level: 90%, 95%, or 99%.
  2. Enter the expected proportion (0 to 1); leave blank to default to 0.5, the most conservative assumption.
  3. Enter the desired margin of error as a percentage.
  4. Select Calculate to get the minimum required sample size.

Related Calculations

For estimating a population mean rather than a proportion, use the Sample Size for Mean Calculator. To check the margin of error for a sample size you already have, see the Margin of Error Calculator.