Monetary unit sampling calculator: formula and example

In monetary unit sampling (MUS) the sample size is n = BV × CF ÷ (TM − EM × EF): population book value times the confidence factor, divided by tolerable misstatement less expected misstatement times the expansion factor. With a €4,812,350 population, €120,000 tolerable misstatement, €15,000 expected misstatement and 95% confidence, you test 151 items.

The calculator uses the factors from the AICPA Audit Sampling guide. Change the figures and the size is recalculated.

MUS sample size calculator

Demo Client Ltd. FY 2026

B-3.1
Confidence factor3.00
Sampling interval€32,000.00
Sample size151 items

PreparedA. Comas,
ReviewedPending
Change the figures: the calculation is real.

What is monetary unit sampling?

Monetary unit sampling (MUS, also called probability-proportional-to-size sampling) is a statistical sampling method in which the sampling unit is each currency unit in the population, not each invoice. An item of €40,000 is forty times more likely to be selected than one of €1,000, so the sample goes where the money is.

It is used mainly for substantive tests of details on balances at risk of overstatement, such as receivables, revenue or inventory. Tests of controls use attribute sampling instead.

MUS sample size formula

n = BV × CF ÷ (TM − EM × EF)

  • BV: population book value. The total you are testing, reconciled to the ledger.
  • CF: confidence factor. Set by the risk of incorrect acceptance you accept: 3.00 for 95% confidence with zero expected errors.
  • TM: tolerable misstatement. The largest misstatement you accept in the population, usually linked to performance materiality.
  • EM: expected misstatement. What you expect to find, based on prior years or your knowledge of the client.
  • EF: expansion factor. Increases expected misstatement to allow for sampling error when you expect misstatements.

The sampling interval is (TM − EM × EF) ÷ CF, which is the same as BV ÷ n. ISA 530 sets no formula: it requires a sample size that reduces sampling risk to an acceptably low level. This is the formula in the AICPA Audit Sampling guide.

Confidence and expansion factors

ConfidenceRisk of incorrect acceptanceConfidence factor (0 errors)Expansion factor
90 %10 %
95 %5 %
99 %1 %

Source: AICPA, Audit Guide: Audit Sampling (Poisson factors).

Monetary unit sampling example

Revenue for a client with a book value of €4,812,350, tolerable misstatement of €120,000, expected misstatement of €15,000 and 95% confidence:

  1. Adjusted expected misstatement: 15,000 × 1.6 = €24,000.
  2. Room left: 120,000 − 24,000 = €96,000.
  3. Sampling interval: 96,000 ÷ 3.00 = €32,000.
  4. Sample size: 4,812,350 ÷ 32,000 = 150.4, rounded up to 151 items.

To select, pick a random start between 1 and 32,000 and take the currency unit that falls in each €32,000 interval of the cumulative listing. Any invoice of €32,000 or more is always selected.

What drives the sample size

With the same €4,812,350 population, this is how the sample changes when you move one input:

ChangeIntervalItems
Base case (95%, TM €120,000, EM €15,000)151
90% confidence115
99% confidence243
No expected misstatement121
Expected misstatement of €30,000201
Tolerable misstatement of €200,00083

As expected misstatement approaches tolerable misstatement, the sample grows very fast. Once EM × EF reaches tolerable misstatement, no sample size works and a different approach is needed.

How the sample is evaluated

For each misstated item smaller than the interval, the misstatement percentage (tainting) is projected over its interval. Items equal to or larger than the interval contribute their actual misstatement. The upper misstatement limit adds projected misstatement, basic precision (CF × interval) and an incremental allowance for each misstatement found. If that limit does not exceed tolerable misstatement, the conclusion is favourable.

When MUS is not a good fit

  • When the risk is understatement: omitted items and zero balances cannot be selected.
  • When the population has many negative or zero balances, which have to be tested separately.
  • When you expect many misstatements: the sample balloons and classical variables sampling with stratification is usually better.

Frequently asked questions

How do you calculate an audit sample size with MUS?

Multiply the population book value by the confidence factor and divide by tolerable misstatement less expected misstatement times the expansion factor. At 95% confidence the confidence factor is 3.00 and the expansion factor 1.6.

What is the sampling interval?

The amount between selections in the cumulative listing of the population: book value divided by sample size. Any item larger than the interval is always selected.

Can you do monetary unit sampling in Excel?

The sample size, yes: it is one formula. Selection over the cumulative listing and evaluation with tainting and an upper limit take more work and are easy to break in a spreadsheet, and the reviewer has to be able to re-perform them.

Which standard covers monetary unit sampling?

ISA 530 internationally, AU-C 530 in the United States and NIA-ES 530 in Spain. None of them requires MUS or a specific formula; the usual factors come from the AICPA Audit Sampling guide.

  • AICPA, Audit Guide: Audit Sampling
  • ICAC, NIA-ES 530 Audit sampling (Spain)
  • IAASB, ISA 530 Audit Sampling

From sample size to workpaper

AuditSampler Pro makes the systematic selection, records the audited value of each item, evaluates the sample with tainting and an upper misstatement limit, and writes the PDF workpaper. It runs on your computer.