Sample Size Calculator
Find the minimum number of samples needed for proportion or mean estimates at your chosen confidence level and precision.
Sample design
Minimum sample size required:
For finite populations, the adjusted sample size uses the finite population correction.
How sample size is determined
The required sample size grows with your desired confidence level and shrinks as you accept a larger margin of error — intuitively, being more certain (higher confidence) or more precise (smaller margin) about a result requires more data. For proportions, the calculation also depends on the estimated proportion itself, since a proportion near 50% requires the largest sample (it has the most variability), while proportions near 0% or 100% require fewer.
Frequently asked questions
What if I don't know the expected proportion in advance? Using 50% is the conservative default — it produces the largest (safest) sample size estimate, so if your actual proportion turns out to be different, your survey still has adequate power.
Why does entering a population size change the result? The finite population correction reduces the required sample size when you're sampling a meaningful fraction of a small, fixed population — for very large or effectively infinite populations, this adjustment makes little difference.
What confidence level should I use? 95% is the most common standard in survey research and social science; 99% is used when higher certainty is needed (e.g., quality control or medical research), at the cost of requiring a larger sample.