Sample Size
Calculator

The standard MUS formula is: n = (Population x Confidence Factor) / (Tolerable Misstatement - Expected Misstatement x Expansion Factor). The confidence factor reflects the acceptable risk of incorrect acceptance. At 90% confidence, the factor is 2.31. Expected misstatement increases the required sample size because the auditor must leave headroom above the expected level before reaching the tolerable threshold.

Items above the sampling interval are automatically selected in full. They form the top stratum. This is because any single item above the interval, if misstated by 100%, would on its own produce a misstatement exceeding the interval, which when projected across the population would exceed tolerable misstatement. These items must be tested individually outside the sample.

Items above the sampling interval are automatically selected in full — they form the top stratum. These items must be tested individually outside the sample. The confidence factor depends on the assessed risk of material misstatement — higher risk means higher confidence required and larger samples. Expected misstatement should be based on prior-period results, understanding of the entity, and results of other audit procedures. Stratification of the population can improve sampling efficiency by reducing variability within each stratum.

Yes. This ISA 530 sampling calculator is completely free, with no login required and no usage limits. All calculations run in your browser. Your population data and audit parameters are never sent to a server. You can use it for every engagement without restriction.

Monetary unit sampling (also called probability-proportional-to-size sampling) is a statistical sampling method where each individual monetary unit in the population has an equal chance of selection (ISA 530.A13). The sampling unit is the individual euro (or dollar), not the transaction. This means larger-value items have a proportionally higher probability of selection, which naturally emphasises the items most likely to cause material misstatement. MUS is the most widely used statistical sampling method for substantive testing of account balances because it automatically stratifies the population by value.

ISA 530.A10 permits either approach, and both can provide sufficient appropriate audit evidence when applied properly. Statistical sampling (including MUS) is preferred when: the population is large and homogeneous, you need to quantify sampling risk precisely, or the engagement requires defensible sample sizes for regulatory inspection. Non-statistical sampling is acceptable when: the population is small, items are heterogeneous and require judgmental selection, or the auditor has strong knowledge of the entity that supports targeted testing. The key requirement is that whichever method you use, the sample must be representative of the population and the results must be evaluated appropriately.

The confidence level reflects the acceptable risk of incorrect acceptance (the risk that the auditor concludes the population is not materially misstated when it actually is). Common levels are 90% (confidence factor 2.31) for moderate-risk populations and 95% (confidence factor 3.0) for higher-risk populations. The choice depends on the assessed risk of material misstatement: if inherent risk and control risk are both high, use a higher confidence level. If you have tested internal controls and found them effective, a lower confidence level may be justified because the overall assurance is supplemented by controls testing.

When you find misstatements in your sample, you must project them to the entire population (ISA 530.A20). For MUS, each misstatement is expressed as a taint (misstatement as a percentage of the item's book value). The projected misstatement is calculated by multiplying the taint by the sampling interval. The total projected misstatement, plus an allowance for sampling risk, is compared against tolerable misstatement. If the total exceeds tolerable misstatement, the auditor concludes the population contains material misstatement and must either extend testing, request management to investigate and correct, or modify the audit opinion.

Yes. When the auditor has tested the operating effectiveness of relevant internal controls and found them effective, the assessed risk of material misstatement is lower. A lower assessed risk means the auditor needs less assurance from substantive procedures, which translates to a lower confidence factor and a smaller sample size (ISA 330.7–9). For example, moving from high assessed risk (confidence factor 3.0) to moderate assessed risk (confidence factor 2.31) can reduce the sample by approximately 23%. However, ISA 330.18 requires that substantive procedures are always performed for material account balances regardless of the control assessment.

The sampling interval is calculated as: Population Value ÷ Sample Size (or equivalently, Tolerable Misstatement ÷ Confidence Factor when expected misstatement is zero). It determines which items fall into the top stratum (individually significant items) and which are sampled. Any item with a book value above the sampling interval must be tested in full. The interval is also used to project misstatements: each tainted item's misstatement is projected across the interval to estimate the population-level effect. A smaller interval means more items are tested, providing greater assurance but at higher cost.

Ciferi offers 20+ free audit and accounting tools, all browser-based with no login required. These include the Materiality Calculator (ISA 320), ISA 570 Going Concern Checklist, Analytical Review Tool (ISA 520), Financial Ratio Calculator, Depreciation Calculator (IAS 16), IFRS 16 Lease Calculator, IFRS 9 ECL Calculator, IAS 37 Provision Calculator, IAS 12 Deferred Tax Calculator, IAS 36 Impairment Calculator, Transfer Pricing Tool, and Intercompany Elimination Tool (IFRS 10). Visit our free tools hub at ciferi.com/free to see the full collection.