Sample Size Calculator
Work out how many responses you need for a reliable survey result.
Why sample size matters
Too few responses give results that jump around. Too many waste time and money. This calculator finds the smallest sample that meets the accuracy you ask for.
Example
For a survey with 95% confidence, a 5% margin of error and the cautious 50% proportion, you need 385 responses for a large population. To estimate an average with a standard deviation of about 10 and a margin of error of 2, you need 97.
Things to keep in mind
- The result assumes a random sample. A biased sample stays biased however large it is.
- Expect some people not to reply, so invite more than the number you need.
- For small populations, enter the population size to reduce the sample you need.
Frequently asked questions
What margin of error should I choose?
Five percent is common for surveys. Smaller margins need many more responses.
Why use 50% as the expected proportion?
It is the most cautious case. Any other proportion needs a smaller or equal sample.
Last updated