Statistical vs. Nonstatistical Audit Sampling Under PCAOB Standards
Auditors rarely examine every transaction in an account or business process. Instead, they frequently rely on audit sampling to obtain sufficient appropriate audit evidence.
But how should the sample be selected?
Should the auditor use a statistical approach, or can professional judgment be used to develop a nonstatistical sample?
Under PCAOB AS 2315: Audit Sampling, both approaches are permitted.
The important issue is not simply whether the sample is statistical or nonstatistical. The real question is whether the auditor has designed, selected, performed, and evaluated the sample in a manner that provides sufficient audit evidence to support the audit conclusion.
What Is Audit Sampling?
PCAOB AS 2315 defines audit sampling as applying an audit procedure to less than 100 percent of the items within an account balance or class of transactions to evaluate a characteristic of that population.
Sampling may be used in both:
Substantive tests of details
Tests of controls
The auditor must consider the audit objective, characteristics of the population, acceptable sampling risk, tolerable misstatement or deviation rate, and other audit evidence when designing the sample.
Statistical Audit Sampling
Statistical sampling uses probability theory to help the auditor design the sample and evaluate its results.
Two characteristics are particularly important:
Random selection. Items are selected using a method that provides a measurable probability of selection.
Mathematical evaluation. Probability theory allows the auditor to quantify sampling risk and evaluate the sample results.
The PCAOB identifies several benefits of statistical sampling. It can help the auditor design an efficient sample, measure the sufficiency of the evidence obtained, and evaluate the results while quantifying sampling risk.
Statistical sampling can therefore provide a disciplined and defensible connection between audit risk, sample size, selection, and evaluation.
Nonstatistical Audit Sampling
Nonstatistical sampling also uses audit sampling—but without using statistical theory to quantify sampling risk.
The auditor instead applies professional judgment when determining matters such as sample size and evaluating results.
That does not mean the auditor can simply pick whatever transactions appear convenient.
PCAOB AS 2315 requires sample items to be selected so that the sample can be expected to be representative of the population. Items within the population should have an opportunity to be selected.
A haphazard selection method may be appropriate in certain nonstatistical applications, but haphazard does not mean careless.
Statistical vs. Nonstatistical: Which Is Better?
PCAOB AS 2315 does not declare one method universally superior.
Either statistical or nonstatistical sampling can provide sufficient audit evidence when properly applied.
The difference is largely in how the auditor designs the sample, measures sampling risk, and evaluates the results.
Statistical Sampling | Nonstatistical Sampling |
Uses probability theory | Relies more heavily on professional judgment |
Quantifies sampling risk | Sampling risk is evaluated judgmentally |
Supports mathematically determined sample sizes | Sample size is determined using professional judgment and relevant sampling factors |
Generally requires random-based selection | May use random or appropriate haphazard selection |
Results evaluated statistically | Results evaluated using professional judgment |
Requires statistical knowledge and methodology | Generally easier to administer |
Neither approach eliminates the need for professional judgment.
Statistics do not replace auditor judgment, and professional judgment does not eliminate sampling risk.
A Common Misunderstanding About Nonstatistical Sampling
One of the most important provisions in AS 2315 concerns sample size.
When circumstances are similar, the factors affecting sample size should have a similar effect regardless of whether statistical or nonstatistical sampling is used.
Consequently, a properly applied nonstatistical sample ordinarily should be comparable to, or larger than, the sample size resulting from an efficient and effectively designed statistical sample.
This is important.
An auditor cannot justify a small sample simply by labeling the approach "nonstatistical."
Professional judgment must have a reasonable basis.
Sampling Risk Still Exists
Whenever auditors examine less than 100 percent of a population, they accept sampling risk.
Sampling risk is the possibility that the auditor's conclusion based on the sample differs from the conclusion that would have been reached if the entire population had been tested.
For substantive testing, an especially important risk is incorrect acceptance—the risk that the sample indicates an account is not materially misstated when it actually is materially misstated.
For tests of controls, the auditor must consider the risk of assessing control risk too low.
PCAOB AS 2315 makes an important relationship clear: As sample size decreases, sampling risk increases.
Sample Selection Matters
An auditor cannot simply select the easiest transactions to test.
AS 2315 requires samples to be selected so they can be expected to be representative of the population.
Depending upon the circumstances, selection techniques can include:
Random selection
Systematic selection with an appropriate random start
Probability-proportional-to-size techniques
Stratified random sampling
Haphazard selection for appropriate nonstatistical applications
The auditor should also separately consider items that warrant 100 percent examination because of their size, risk, unusual characteristics, or potential for material misstatement.
Those items are not part of the remaining population being sampled.
What Happens When the Auditor Finds Errors?
Selecting and testing the sample is only part of the process.
The auditor must evaluate the results.
For substantive tests of details, AS 2315 requires the auditor to project sample misstatements to the population from which the sample was selected.
The auditor must also consider qualitative factors.
An error caused by an isolated clerical mistake may have very different implications from a misstatement involving:
Management override
Fraud
An unusual journal entry
A control failure
Management integrity
A systematic processing problem
A discovered misstatement may also indicate that the assumptions used to design the audit were incorrect.
That could require the auditor to reconsider the risk assessment or perform additional procedures.
Audit Sampling Is More Than Picking 25 Transactions
A weak approach to audit sampling often begins with a predetermined sample size: "We always test 25."
That reverses the proper audit logic.
The auditor should first determine:
What is the audit objective?
What assertion is being tested?
What is the population?
What is the risk of material misstatement?
What is tolerable misstatement or the tolerable deviation rate?
What other audit evidence is available?
What level of sampling risk is acceptable?
Only then should the auditor determine an appropriate sample size and selection methodology.
Sample size should be the result of audit planning—not the starting point.
The Bottom Line
PCAOB AS 2315 permits auditors to use either statistical or nonstatistical audit sampling.
Both can provide sufficient audit evidence when properly designed and applied.
Statistical sampling gives the auditor the ability to quantify sampling risk using probability theory. Nonstatistical sampling relies more extensively on professional judgment.
But nonstatistical does not mean informal.
The auditor still needs a defensible relationship among:
Audit Objective → Risk → Population → Sample Size → Selection → Testing → Evaluation → Audit Conclusion
The workpapers should tell that story.
If they do not, the problem may not be the sampling technique.
The problem may be the audit methodology.
Learn PCAOB Audit Sampling with Corporate Compliance Seminars
Corporate Compliance Seminars provides CPE programs addressing PCAOB Auditing Standards and audit sampling, including training focused specifically on PCAOB AS 2315: Audit Sampling.
CCS PCAOB training helps external auditors move beyond simply reading the standard and understand how PCAOB requirements affect actual audit planning, workpaper documentation, substantive testing, tests of controls, audit evidence, risk assessment, and inspection readiness.
Other CCS PCAOB CPE programs address interconnected standards involving:
AS 1105 — Audit Evidence
AS 2101 — Audit Planning
AS 2301 — The Auditor's Responses to the Risks of Material Misstatement
AS 2810 — Evaluating Audit Results
PCAOB audit documentation
PCAOB inspection findings
Broker-dealer auditing
PCAOB firm quality control
For auditors working on SEC issuer or registered broker-dealer engagements, understanding sampling is not simply about knowing how many transactions to select.
It is about being able to demonstrate why the sample provides sufficient appropriate audit evidence to support the auditor's conclusion.
That is the standard the audit workpapers ultimately need to meet.
Comments