How many respondents do you need?
Compute the sample size for your survey, the margin of error of an existing sample, or a confidence interval for a mean or proportion — including finite population correction.
With 385 completed responses, your estimate is within ±5% of the true value in 95% of samples.
Formulas assume (simple) random sampling. Convenience samples, quota designs, and cluster sampling need design-effect corrections these calculators don't apply.
Frequently asked questions
How many respondents do I need for my survey?
The standard answer for a 95% confidence level and ±5% margin of error is 385 respondents — regardless of population size, unless the population is small. Enter it above and the finite population correction shrinks the number.
What is the 50% default for the expected proportion?
p = 50% maximizes variance and therefore gives the most conservative (largest) sample size. If prior research suggests the true proportion is nearer 20% or 80%, enter that and the required sample shrinks.
What does the margin of error mean?
It is the half-width of the confidence interval around your estimate: a survey result of 60% with a ±4% margin means the true value plausibly lies between 56% and 64% at your chosen confidence level.
Does a bigger population need a bigger sample?
Barely — above roughly 20,000 people the required sample is essentially constant. Population size only matters for small populations, where the finite population correction reduces the required n.
More free tools
Need to check the whole paper?
Scan your citations and AI-generated passages for free, pay only to unlock the full report.
Scan your thesis free