Sample Size Calculator

What this tool does

Sample Size Calculator determines the minimum number of survey respondents needed for a target confidence level and margin of error, with an automatic finite population correction applied when you specify a known, bounded population size.

Designing a survey means deciding in advance how many respondents you actually need, too few and your results carry too much uncertainty, too many and you’ve spent more time and resources than necessary.

This calculator determines that required sample size from your target confidence level, acceptable margin of error, and an optional population size, with the finite population correction applied automatically when you specify a known, bounded population.

How to use Sample Size Calculator

1. Choose your target confidence level (90%, 95%, or 99%).
2. Set your acceptable margin of error.
3. Enter an estimated response distribution, or leave at 50% if unknown.
4. Enter your population size if known and bounded, or leave at 0 for a large/unlimited population.

Tips for better results

• Using 50% for the estimated response distribution is the safest, most conservative default when you don’t already know roughly how responses will split, it produces the largest required sample size, erring on the side of more data rather than less.
• A 95% confidence level with a 5% margin of error is a very common standard for general surveys, a reasonable default if you’re not sure what to use.
• The finite population correction only matters meaningfully when your population is relatively small (in the hundreds or low thousands); for a large population (tens of thousands or more), leaving population at 0 produces essentially the same required sample size.
• A smaller margin of error requires a substantially larger sample size, the relationship isn’t linear, halving your margin of error roughly quadruples the required sample.

Common use cases

• Planning how many respondents a survey needs before launching it.
• Determining an appropriate sample size for a research study or poll.
• Checking whether an existing sample size is large enough for a desired confidence level and margin of error.
• Understanding the tradeoff between survey precision (margin of error) and required respondent count.

Common mistakes to avoid

• Guessing a response distribution far from 50% without good evidence, this can meaningfully under-estimate the required sample size if your guess turns out to be wrong, 50% is the conservative default precisely because it doesn’t require a confident prior estimate.
• Forgetting the finite population correction only matters for a relatively small, known population, entering an unnecessarily small population figure when your actual target audience is effectively unlimited will produce a misleadingly small required sample.
• Expecting a linear relationship between margin of error and sample size, halving the margin of error roughly quadruples the required sample size, not just doubles it.