When a test of controls or a test of details looks at part of a population, the auditor has three ways to choose the items: examine all of them, select specific items, or sample. The requirement is in AU-C 330, paragraph .25: the auditor determines the means of selecting items that are effective in meeting the purpose of the procedure. The detail is in application material A68 to A74. AU-C 500 supplies the companion rule that the population itself must be reliable before anything is selected from it.
This guide covers how the three means differ, what each lets you conclude, and how to combine them. For the sampling mechanics see audit sampling methods, and for the evidence rules behind it see audit evidence under AU-C 500.
What are the three means of selection?
A68 lists them:
- selecting all items (100 percent examination);
- selecting specific items; and
- audit sampling.
A69 says any one or a combination may be appropriate depending on the risks of material misstatement for the assertion and the practicality and efficiency of each. The test of effectiveness is not only relevance and reliability, which AU-C 500 paragraph .07 requires the auditor to consider, but also extent: how much of the population the procedure covers (A68).
When is 100 percent examination appropriate?
A70 says the auditor may decide to examine the entire population, or a stratum within it. It is unlikely for tests of controls and more common for tests of details. Three examples are given:
- the population is a small number of large-value items;
- a significant risk exists and other means do not provide sufficient appropriate evidence; or
- the repetitive nature of a calculation or process run by an information system makes 100 percent examination cost effective.
The third is where automated tools earn their keep. AU-C 500 A61 describes an auditor using automated tools to scan an entire population and identify transactions meeting criteria for being unusual. Items that show characteristics of risk are then examined, and the auditor determines that items not selected are less likely to be materially misstated. A61 says that analytical work provides evidence about both groups of items. See audit data analytics and full-population testing.
When do you select specific items?
A71 says the decision draws on the auditor's understanding of the entity, the assessed risks of material misstatement, and the characteristics of the population. The standard lists three kinds of specific items:
- High-value or key items. Items that are high in value or have another characteristic: suspicious, unusual, risk-prone, or with a history of error.
- All items over a certain amount. Used to verify a large proportion of the total amount of a class or balance.
- Items selected to obtain information. For example, to learn about the entity or the nature of its transactions.
A71 also says judgmental selection is subject to nonsampling risk. The bigger limit is in A72: selecting specific items can be efficient, but it does not constitute audit sampling. The results cannot be projected to the entire population, and examining specific items does not, by itself, provide sufficient appropriate audit evidence about the remainder of the population.
That sentence drives the file. Key items are covered by their own results. The remaining items need another response: a sample, analytical procedures, or a documented conclusion that the remainder is not material enough to need any.
When do you sample?
A73 describes audit sampling as designed to enable conclusions about an entire population on the basis of testing a sample drawn from it, and points to AU-C 530. A74 warns that valid conclusions ordinarily can be drawn using sampling, but an unacceptable risk exists if the sample is too small, the approach or selection method does not fit the objective, or exceptions are not followed up.
AU-C 530 paragraph .08 requires a selection method that lets the auditor reasonably expect a representative sample. Its A15 describes the combined approach directly: the auditor might first separately examine items deemed of relatively high risk, then use sampling to form an estimate of some characteristic of the remaining population.
What does AU-C 500 add about the population?
Paragraph .08 requires the auditor's evaluation of information used as audit evidence to include whether it is sufficiently precise and detailed and, as necessary, evidence about its accuracy and completeness. A40 gives an example that applies straight to selection: a population being tested for a certain characteristic, such as payment authorization, is affected by the completeness of the population from which items are selected. AU-C 530 A8 repeats that the auditor performs procedures to obtain evidence that the population drawn on is complete.
Practically, tie the listing you select from to the general ledger or trial balance before choosing a single item. A sample of disbursements drawn from a report that omitted a bank account supports nothing about that account.
Direction matters too. AS 2315 paragraph .17 gives the PCAOB version: recorded payables cannot reveal omitted ones, so to test for unrecorded items the auditor selects from a source in which they would appear, such as subsequent cash disbursements.
Worked example: accounts payable invoices
A population of 244 vendor invoices recorded in the last quarter totals $3,200,000. Performance materiality is $120,000.
Step 1: key items (specific selection). Nine invoices exceed $100,000, totaling $1,450,000. The auditor adds four more because of risk characteristics: round-dollar amounts near year-end and one related party, totaling $120,000. That is 13 invoices and $1,570,000, which is 49.1 percent of the dollar value ($1,570,000 / $3,200,000).
Step 2: sample the rest. The remaining 231 invoices total $1,630,000 ($3,200,000 minus $1,570,000). The auditor selects 21 using systematic selection with a random start: 231 / 21 = an interval of 11 items.
Step 3: results.
- Key items: $4,000 overstatement found. Under A72, it is a known misstatement and is not projected.
- Sample: 21 items with a book value of $151,300 show a $2,900 overstatement. The misstatement rate is $2,900 / $151,300 = 1.92 percent. Projected to the remaining population: 1.92 percent of $1,630,000 is about $31,243.
Step 4: evaluate. Known $4,000 plus projected $31,243 is $35,243, which is 29.4 percent of performance materiality of $120,000. The auditor also investigates the cause (AU-C 530 paragraph .12): if the overstatements share a vendor or a period, the auditor looks at all items with that feature (A22) rather than relying on the projection alone.
The file shows the three groups, why each item is in its group, and that the groups add back to the population total of $3,200,000.
Frequently asked questions
Can haphazard picks count as sampling?
For nonstatistical sampling, A17 of AU-C 530 says haphazard and random selection are the principal techniques. Judgmental picks of "interesting" items are specific item selection and cannot be projected.
Do I need to sample the remainder at all?
Key-item results alone do not support a conclusion about the remainder (A72). If the auditor decides the remainder needs no sample, the file shows what other evidence, such as analytical procedures or low risk characteristics of the remainder, supports that conclusion.
Where do tests of controls fit?
A 100 percent examination is unlikely for tests of controls (A70), so the choice is between specific items and sampling.
Sources
- AICPA, AU-C Section 330, Performing Audit Procedures in Response to Assessed Risks and Evaluating the Audit Evidence Obtained (paragraph .25, A68 to A74): https://www.aicpa-cima.com/resources/landing/aicpa-professional-standards
- AICPA, AU-C Section 500, Audit Evidence (paragraphs .07 to .08, A40, A61): https://www.aicpa-cima.com/resources/landing/aicpa-professional-standards
- AICPA, AU-C Section 530, Audit Sampling (paragraph .08, A8, A15, A17, A22): https://www.aicpa-cima.com/resources/landing/aicpa-professional-standards
- PCAOB, AS 2315, Audit Sampling (paragraph .17): https://pcaobus.org/oversight/standards/auditing-standards/details/AS2315
CORAA prepares ready-to-review audit files for US CPA firms from client books; you review and sign. See /us.
Part of CORAA University · Evidence and audit procedures: related guides, calculators and templates.
AU-C references are to AICPA Professional Standards, paraphrased here; read the standard itself before relying on a requirement. This page is guidance, not professional advice. Report an error