Audit and Assurance · Audit sampling and other means of testing

Audit Sample Selection Methods for ACCA AA

Updated 11 October 2026 · Fact-checked

Sample selection methods are the ways an auditor chooses items to test: random, systematic, monetary unit sampling (MUS), haphazard and block selection. A sample is statistical only if it uses random selection and evaluates results using probability theory. Otherwise it is non-statistical. To answer a question, match the method to the population, then state its advantage and weakness.

Understand Sample Selection Methods

An auditor cannot test every transaction. So the auditor tests a sample and draws a conclusion about the whole population. For the conclusion to be valid, the sample must be representative. That means every item in the population has a chance of being picked, and the pick is not biased.

Selection methods are the tools for choosing items. Random selection gives every item an equal chance. You use random number tables or a computer generator. It is unbiased, but you need a complete, numbered population, and it can be slow to find the items.

Systematic selection picks every nth item after a random start. You work out the interval by dividing the population size by the sample size. It is simple and quick. The risk is that the population has a pattern that matches the interval, for example every 10th invoice being a month-end one. That would bias the sample.

Monetary unit sampling (MUS) treats each ₹1 (or $1) in the population as a sampling unit. Larger items have a greater chance of being picked, because they contain more units. It is good for testing overstatement, such as receivables or inventory. Its weakness is that understated or zero-value items are unlikely to be picked, so it is poor for testing understatement, like unrecorded liabilities.

Haphazard selection means picking items with no structure, trying to avoid bias. It is not statistical. The danger is that the auditor, even without meaning to, favours some items, such as those easy to reach. Block selection means picking a contiguous group, such as all invoices in one week. It is easy, but a block is rarely representative. ISA 530 Appendix 4 notes that block selection is normally not appropriate for audit sampling, since valid inferences about the whole population are rarely possible from contiguous blocks, even with several blocks.

Random, systematic, MUS, haphazard and block selection are all ways of choosing items. None of them is statistical in itself. A sample is statistical only if it combines two things: random selection of the sample items, and evaluation of the results using probability theory. Systematic selection with a random start and MUS can be used in a statistical sample when the results are evaluated using probability theory. Random selection can also be used in a non-statistical sample if the results are not evaluated using probability theory. A statistical sample allows the auditor to measure sampling risk mathematically. Non-statistical sampling relies on judgement. Both can give sufficient appropriate evidence if designed well. ISA 530 requires the auditor to select sample items so that all sampling units in the population have a chance of selection. This is why block selection is normally not appropriate for audit sampling.

Key rules to remember

Sampling interval (systematic): Interval = Population size ÷ Sample size; Choose a random start between 1 and the interval, then take every nth item.
Sampling interval (monetary unit sampling): Interval = Total population value ÷ Sample size; With fixed-interval selection, items with a value at or above the interval are certain to be selected.
Probability of selection in MUS: Probability = Item value ÷ Sampling interval, for items below the interval; This is exact with fixed-interval selection from a random start. It explains why larger items are more likely to be picked. Items equal to or above the interval are certain to be selected.
Statistical sampling test: Statistical sample = random selection of items + evaluation of results using probability theory; Systematic selection with a random start and MUS can be used in a statistical sample when results are evaluated using probability theory. Selection alone does not make a sample statistical. Haphazard and block selection are non-statistical.

How to solve Sample Selection Methods questions

Use this method for any question asking you to choose, explain or compare selection methods.

1Read the scenario and identify the population: what is being tested, how many items, and their values.
2Identify the assertion or risk. Overstatement points to MUS. Completeness or understatement points away from MUS.
3Check what is available: a complete, numbered list suits random or systematic selection.
4Choose the method and say exactly how you would apply it, for example the interval and a random start.
5State one advantage that fits the scenario, such as unbiased, quick or focuses on high values.
6State one weakness or risk that fits the scenario, such as pattern bias or poor for understatement.
7Say whether the method is statistical, and whether you should add key items or specific items selected separately.
8Finish with a short conclusion that links the method to sufficient appropriate evidence.

Quickest way: Match method to scenario in 30 seconds

When to use it: Use this for Section A and Section B objective questions where you must pick the best method or spot a flaw.

Overstatement of high-value balances: choose monetary unit sampling.
Numbered documents and a need for speed: choose systematic, check for pattern risk.
Need for an unbiased, defensible sample: choose random.
No structure and a claim of no bias: this is haphazard, which is non-statistical.
One group such as a single month: this is block selection, which is rarely representative.
If the option says a method guarantees representativeness, reject it.

Common mistakes in Sample Selection Methods

Calling haphazard selection a statistical or random method.
The word 'haphazard' sounds like random.
Fix: Random means every item has a known chance, usually through a number generator. Haphazard is judgemental and non-statistical.
Using monetary unit sampling to test understatement or completeness of liabilities.
Students remember MUS as the 'best' method.
Fix: MUS favours large items and misses small or zero items. Use it for overstatement. For completeness, test from outside the recorded population.
Forgetting the pattern risk in systematic selection.
The method seems simple, so students see only the advantages.
Fix: Always state that a pattern in the population matching the interval can bias the sample, and say you use a random start.
Saying block selection is acceptable because it is random.
Students think any group chosen without thought is fair.
Fix: Block selection is not random. A block is one contiguous group and does not give all items a chance of selection. ISA 530 Appendix 4 notes that block selection is normally not appropriate for audit sampling, since valid inferences about the whole population are rarely possible from contiguous blocks, even with several blocks.
Calculating the interval with the wrong figures.
Students mix up population size with population value.
Fix: Systematic uses item count. MUS uses total monetary value. Write the units next to each number.
Giving only advantages or only disadvantages.
Students rush and forget that questions ask for both.
Fix: Give one point each side for every method, linked to the scenario.

Worked examples

Example 1

An audit team will test 50 sales invoices from a population of 2,000 numbered invoices using systematic selection. Calculate the sampling interval, say which invoices are tested if the random start is 17, and state one weakness of the method. (4 marks)

Show the solution

Interval = 2,000 ÷ 50 = 40.
The first invoice is number 17, from the random start.
Add 40 each time: 17, 57, 97, 137 and so on, up to 1,977.
Check the count: invoice numbers 17 + 40 × 49 = 1,977, which is the 50th item, so there are 50 items.
Weakness: if the invoices follow a pattern with a period that matches 40, for example one type of invoice appearing every 40th, the sample would be biased.

Answer: The interval is 40. Invoices 17, 57, 97, 137 and so on up to 1,977 are tested. The weakness is that a pattern in the population matching the interval can make the sample unrepresentative.

Example 2

A client has a receivables ledger with a total value of ₹6,00,00,000. The auditor wants to test for overstatement using monetary unit sampling with a sample size of 40. Calculate the sampling interval. A customer balance is ₹20,00,000. Explain whether it will be selected, and explain why monetary unit sampling is not suited to testing completeness of payables. (5 marks)

Show the solution

Interval = ₹6,00,00,000 ÷ 40 = ₹15,00,000.
The balance of ₹20,00,000 is larger than the interval of ₹15,00,000.
With fixed-interval selection, an item with a value at least equal to the interval must contain at least one selected monetary unit, so it is certain to be selected.
Completeness of payables is about understatement, such as unrecorded liabilities.
An unrecorded liability has no value in the population, so it has no chance of being selected. Small items are also unlikely to be picked.

Answer: The interval is ₹15,00,000. The ₹20,00,000 balance is above it, so it will certainly be selected. MUS is unsuitable for completeness of payables because unrecorded or understated items have little or no chance of selection.

Exam tips

Always state whether the method is statistical or non-statistical. Markers look for this.
For every method, write one advantage and one disadvantage. Link them to the scenario given.
In calculation questions, show the interval formula first, then the working. Method marks are available in Section C.
In objective questions, watch for options that overstate a method, such as 'guarantees' or 'eliminates sampling risk'. These are wrong.
Remember that selecting items is separate from sampling: key items and high-value items may be tested outright, with sampling applied to the rest.

Sample Selection Methods: frequently asked questions

What is the difference between random and systematic sampling in audit?

Random selection picks items using random numbers, so every item has an equal chance. Systematic selection picks every nth item after a random start. Systematic is quicker but can be biased if the population has a pattern.

What is monetary unit sampling in ACCA AA?

Monetary unit sampling treats each unit of currency in the population as a sampling unit. Larger items are more likely to be selected. It suits testing overstatement but is weak for understatement.

What is the difference between haphazard and block selection?

Haphazard selection picks items with no set structure, aiming to avoid bias. Block selection picks a contiguous group, such as all items in one week. Both are non-statistical. ISA 530 Appendix 4 notes that block selection is normally not appropriate for audit sampling, since valid inferences about the population are rarely possible from contiguous blocks.

Is haphazard selection statistical?

No. It is non-statistical, so it does not allow sampling risk to be measured mathematically. It can be used in non-statistical sampling if the auditor tries to avoid bias or predictability and selects items without regard to their size, source or other characteristics. ISA 530 says it is not appropriate when using statistical sampling.

How do I answer a sample selection question in the exam?

Identify the population and the risk, name the method, and explain how it would be applied. Then give an advantage and a weakness. Link each point to the scenario.