Estimate a proportion-based audit sample for a finite population of maintenance records. Choose the population size, confidence level, margin of error, expected exception proportion, and a reserve for unusable selections. The calculator reports the base statistical sample, the reserve-adjusted number to select, the sampling fraction, and the approximate precision achieved after rounding. This is a planning model; it does not choose the audit objective or replace a required professional methodology.
N is population size, z is the selected normal critical value, p is expected proportion, and e is margin of error as a decimal.
The base sample is the finite-population estimate before a reserve. Adjusted selections are how many records to draw so the usable sample can still meet the base target if the entered reserve rate materializes.
This calculator estimates a simple random sample for a proportion. It does not address monetary-unit sampling, stratification, discovery sampling, control reliance, clustered data, or mandatory audit standards.
For 5,000 records at 95% confidence, 5% margin of error, and a 50% expected proportion, the finite-population base sample rounds up to 357 records. With a 10% reserve, select 397 records.
When the population is not extremely large, sampling without replacement reduces the sample needed for the selected confidence and margin of error. The correction accounts for that relationship.
Use 50% for the most conservative proportion-based sample under this model. A supported prior estimate may produce a smaller sample.
No. Sampling design, population definition, selection method, evidence quality, and applicable audit requirements also matter.
Yes. The reserve percentage inflates the statistical sample, and the calculator reports both the base and adjusted sample.
No. The appropriate confidence and precision depend on the audit objective, risk, and governing methodology. Select them deliberately rather than treating one setting as universal.
| Variable | Meaning | Effect when increased |
|---|---|---|
| Confidence level | Selected statistical confidence | Larger sample |
| Margin of error | Target half-width | Smaller sample |
| Expected proportion | Modeled exception share | Largest sample near 50% |
| Reserve | Allowance for unusable items | More selections, same base target |