The famous 385
For ±5% at 95% confidence with an unknown proportion (use 50%): n = 1.96² × 0.25 ÷ 0.05² = 384.16, rounded up to 385. That is why so many surveys aim for about 400 responses. For ±3% you need 1,068.
Small populations
If the population is small, you need fewer responses. For a school of 600 students, the 385 above becomes 385 ÷ (1 + 384 ÷ 600) ≈ 235. Enter the population size to apply this correction.
For a mean, you need an estimate of σ, from a pilot study, earlier research, or range ÷ 4 as a rough guess.
Frequently asked questions
Why always round up?
Rounding down would give a margin of error slightly larger than your target. You can’t survey a fraction of a person, so take the next whole number.
Does this work for hypothesis tests?
This calculator sizes a study for estimation precision. Sizing for a hypothesis test also needs a target effect size and power (often 80%), which is a different calculation.