Sample size calculator

Find out instantly how many people you need to survey for your research, thesis or poll.

Your data

🔒 Your data is processed in your browser and never sent to a server.

Result

278

people minimum

Rounded up to keep the margin of error you chose.

Z value1.96
Uncorrected sample384.1

What is sample size and why calculate it?

It is the minimum number of people, records or observations you need to study to estimate a population characteristic with the precision and confidence you choose. Calculating the right sample size saves time and money without losing rigor, and it is a common requirement in surveys, theses and research projects. This calculator uses the simple random sampling formula for a proportion.

How to calculate sample size step by step

  1. Enter your population size. If it is very large or unknown, leave it at 0.
  2. Pick a confidence level; 95% is the most common.
  3. Set the margin of error you can accept, for example 5%.
  4. Use an expected proportion of 0.50 if you have no prior estimate.

Sample size formula (finite and infinite population)

n₀ = Z² · p(1−p) / e² n = n₀ / [1 + (n₀−1)/N]

Z is the critical value for the confidence level, p the expected proportion, e the margin of error and N the population size. The second formula applies the finite population correction.

Example

For a population of 1,000 people, 95% confidence, a 5% margin of error and p = 0.50, you need to survey 278 people. For an infinite population it would be 385.

Frequently asked questions

How do I calculate the sample size for a survey?

Define your population, choose 95% confidence and a 5% margin of error, and use p = 0.50 if you have no prior estimate. The calculator returns the minimum number of responses you need.

What is the difference between a finite and an infinite population?

A population is finite when you know how many members it has (for example, 1,000 employees), and the formula applies a correction that lowers the sample. If it is very large or its size is unknown, treat it as infinite.

Which margin of error should I use?

5% is the most common choice. A smaller margin gives more precision but requires a larger sample.

What if I do not know the expected proportion?

Use 0.50. It is the most conservative choice because it gives the largest sample for the other settings.

Does it work for stratified or cluster sampling?

This calculator assumes simple random sampling. For stratified and cluster sampling, try our Android app.

More calculators