Sample Size Calculator

Find how many survey responses or test subjects you need for a given confidence level and margin of error — with optional finite population correction.

816 views

How Many Responses Do You Need?

The required sample size is n = Z² × p × (1−p) / e², where Z comes from your chosen confidence level (90/95/99%), p is the expected proportion (50% is the safest, most conservative assumption when you don't know it), and e is your margin of error. If you know the total population size, a finite population correction shrinks the required sample further — smaller populations need proportionally fewer responses.

Frequently Asked Questions

Why default to 50% expected proportion?

50% (p=0.5) maximizes p×(1−p), which gives the largest, most conservative required sample size — safe when you have no prior estimate of the true proportion.

Do I need the population size?

Only if it's relatively small (a few thousand or less) and known. For large or unknown populations (e.g. "all adults in a country"), leave it blank — the standard formula already assumes an effectively infinite population.

What confidence level should I pick?

95% is the most common default for surveys and research. 99% is stricter (larger sample, more certainty); 90% is looser (smaller sample, less certainty).

Similar Tools