Sample Size Calculator
Calculate the required sample size for surveys and studies based on confidence level, margin of error, and population size.
Required Sample Size
—
Infinite Population n
—
Finite Correction Applied
—
What is Sample Size Calculator?
The sample size calculator determines how many participants you need in a study or survey to achieve statistically valid results at your chosen confidence level and margin of error. It supports finite population correction for known population sizes.
How to use
- 1 Select your desired confidence level (90%, 95%, or 99%).
- 2 Enter the margin of error as a percentage (e.g., 5 for ±5%).
- 3 Enter the population proportion if known (default 50% gives the most conservative estimate).
- 4 Optionally enter the total population size for finite population correction.
- 5 Click Calculate to see the required sample size.
Formula
Example calculation
95% confidence, ±5% margin, 50% proportion: n = 1.96² × 0.5 × 0.5 / 0.05² = 384.16 ≈ 385. With population N=1000: n_adj = 385 / (1 + 384/1000) ≈ 278.
Frequently asked questions
Why is 50% the default proportion?
Using p=50% (0.5) maximizes p×(1−p), which gives the largest and most conservative sample size estimate. If you have prior data on the proportion, use that for a more precise estimate.
What confidence level should I use?
95% is the most common standard in research and business. 99% is used when errors are costly (medical, financial). 90% is acceptable for preliminary or low-stakes research.
What is finite population correction?
When your sample is a significant fraction of a small population, you can reduce the required sample size using the finite population correction factor. Enter the total population size to apply it.
What margin of error is acceptable?
±5% is the standard for most surveys. ±3% is used when precision is critical (political polls, medical studies). ±10% is acceptable for exploratory research.
Does sample size affect accuracy?
Yes. Larger samples reduce the margin of error and increase confidence in your results. However, returns diminish rapidly — doubling precision requires quadrupling the sample size.