What the Sample size works out
Give the Sample size the values it asks for and it reports sample size, worked out from the formula n₀ = z²p(1−p)/e²; finite population n = n₀/(1+(n₀−1)/N). For estimating a population proportion with a simple random sample.
What each field means
- Confidence (%)
- A yearly percentage typed as a plain number, so 95 means 95%. Values must fall between 50 and 99.99.
- Margin of error (%)
- A yearly percentage typed as a plain number, so 5 means 5%. Values must fall between 0.001 and 99.
- Expected proportion (%)
- A yearly percentage typed as a plain number, so 50 means 50%. Values must fall between 0.001 and 99.999.
- Population size (0 = unlimited)
- A number; the example uses 0. Values must fall between 0 and 1000000000.
A worked example
With the values this page opens on — Confidence (%) 95, Margin of error (%) 5, Expected proportion (%) 50, Population size (0 = unlimited) 0 — the answer is 385. The supporting detail reads: Responses needed. Reset restores exactly these numbers.
Reading the result
What you get back is sample size. Each edit recomputes the answer and redraws the chart from your own figures, so the displayed rounding never feeds back into the calculation.
Assumptions and limits
For estimating a population proportion with a simple random sample. Does not account for design effects or nonresponse.
Your privacy
Nothing you type leaves your browser.
Related calculators
Tire size ↗ · Shoe size estimate ↗ · Data size ↗ · Bra size starting estimate ↗