Sample size calculator
Find how many units to inspect for a given confidence level and margin of error.
How it is calculated
n₀ = Z² × p × (1 − p) ÷ E². With a known population: n = n₀ ÷ (1 + (n₀ − 1) ÷ N). Round up. Z = 1.645, 1.96 or 2.576.
Worked example
95 %, p = 0.5, E = 0.05: n₀ = 384.16, so 385 units.
Steps
- Choose the confidence level.
- Enter p, or 0.5 if you have no estimate.
- Enter lot size if known.
Frequently asked questions
Why p = 0.5?
It gives the largest sample, so it is safe when you do not know the defect rate.
What is a finite population correction?
It reduces the sample for small lots, where sampling without replacement matters.
Is this an AQL plan?
No. For AQL use ISO 2859-1 sampling tables.
References
Standard proportion sample size formula (Cochran). For acceptance sampling, see ISO 2859-1.
Estimates for information only. Verify critical values with a qualified engineer or your supplier standards.