Sampling and documentation - Statistical vs non-statistical ...

Learning Outcomes

After reading this article, you will be able to identify when and why auditors use sampling, clearly distinguish between statistical and non-statistical sampling in the context of ISA 530, describe the documentation requirements for sampling in audits, and explain sampling risk, stratification, and result evaluation. You should be able to select and justify appropriate sampling techniques in exam scenarios.

ACCA Audit and Assurance (AA) Syllabus

For ACCA Audit and Assurance (AA), you are required to understand sampling as a core part of audit evidence and procedures. Focus your revision on:

• The definition and purpose of audit sampling, including when it is necessary under the ISAs.
• The key differences between statistical and non-statistical sampling, including how samples are selected and results evaluated.
• The requirements of ISA 530: sampling design, size, selection, performance, and evaluation.
• Understanding sampling risk and non-sampling risk in relation to detection risk.
• Factors that affect sample sizes for audit testing.
• Methods of selecting sample items: random, systematic, haphazard, block, and monetary unit sampling.
• Documentation obligations over sampling approach, sample selection, and evaluation of deviations and misstatements.

Test Your Knowledge

Attempt these questions before reading this article. If you find some difficult or cannot remember the answers, remember to look more closely at that area during your revision.

1. Explain what is meant by audit sampling and give one practical reason why sampling is often used in audits.
2. Identify two key features that distinguish statistical from non-statistical sampling.
3. Describe how sampling risk can affect an auditor’s conclusions.
4. State one ISA 530 documentation requirement relating to sampling.

Introduction

Audit sampling is used when it is not practical or cost-effective to examine all items in a population. ISA 530 sets out how sampling should be planned, performed, and documented. Understanding both statistical and non-statistical sampling approaches is critical, as is the ability to select, document, and evaluate audit samples appropriately for ACCA AA.

Key Term: audit sampling
The application of audit procedures to less than 100% of items within a population so that all sampling units have a chance of selection, providing a reasonable basis to draw conclusions about the population as a whole.

What is Audit Sampling?

Audit sampling enables the auditor to form conclusions about an account balance or class of transactions without testing every item.

Audit sampling is only suitable where the population is sufficiently large and homogeneous, and the auditor expects similar characteristics across items.

Key Term: sampling risk
The risk that the auditor’s conclusion based on a sample differs from the conclusion if the entire population were tested (e.g., concluding that a control is effective when it is not).

Key Term: non-sampling risk
The risk that the auditor’s conclusion is erroneous for reasons unrelated to sample size or selection, such as mistakes in execution, interpretation, or judgement.

Sampling Methods

There are two broad methods for audit sampling under ISA 530:

Statistical Sampling

Statistical sampling uses random selection and probability theory to evaluate the results. It produces an objectively measurable basis for conclusions.

Key Term: statistical sampling
An approach to audit sampling that uses random selection and probability theory for sample evaluation, giving each item a known chance of selection and quantifying sampling risk.

Non-Statistical Sampling

Non-statistical sampling (also called judgemental sampling) uses auditor judgement for sample selection and result evaluation. There is no formal measurement of sampling risk.

Key Term: non-statistical sampling
Sampling that does not use statistical techniques or probability theory, relying instead on auditor judgement to select samples and evaluate findings.

Sampling Selection Techniques

Whether using statistical or non-statistical methods, selection should allow all items a chance of selection, except when targeting specific risk areas.

Common selection techniques include:

• Random selection: Every item has an equal chance of being chosen (statistical).
• Systematic selection: Selecting every nth item after a random start (statistical).
• Monetary unit sampling: Stratified random selection weighted by item size (statistical).
• Haphazard selection: Selection based on no specific pattern but aiming to avoid bias (non-statistical).
• Block selection: Testing a contiguous group of items, e.g., all invoices in June—rarely appropriate due to bias risks (non-statistical).

Worked Example 1.1

Scenario:
An auditor wishes to test the effectiveness of purchase order authorisation controls. The client processed 4,000 purchase orders during the year. The auditor decides to test a sample due to time constraints.

Question:
Which sample selection method would allow the auditor to statistically measure sampling risk, and why?

Answer:
Using random or systematic selection, every purchase order has a known probability of selection. Statistical sampling can then be used to compute sampling risk and draw conclusions about the population.

Designing the Sample

ISA 530 requires the auditor to consider:

• Purpose of the procedure (e.g., test controls or substantive testing).
• Materiality and risk in the population.
• Characteristics and size of the population.
• Expected error/deviation rate.
• Tolerable misstatement or deviation.

Sample size increases with higher risk and decreases when controls or other evidence reduce risk.

Key Term: tolerable misstatement
The maximum error in a population that the auditor is willing to accept without concluding the financial statements are materially misstated.

Stratification

Stratification involves dividing a population into sub-groups with similar characteristics to improve sample representativeness and efficiency, often used for high-value or unusual items.

Key Term: stratification
The process of subdividing a population into subgroups of similar items so sample sizes can be reduced and conclusions for each sub-population drawn more effectively.

Performing Audit Procedures

Each item selected must be tested as planned. If an item cannot be tested (e.g., missing document), it should be treated as a deviation or misstatement.

Any identified error must be considered for its effect on the population and the overall audit opinion.

Worked Example 1.2

Scenario:
In testing a sample of 60 sales invoices for accuracy, the auditor finds two with calculation errors. The total value of errors is £400. The population value is £120,000, and tolerable misstatement is £1,500.

Question:
Should the auditor extend testing, or can they conclude the population is free from material misstatement?

Answer:
The projected misstatement is (£400 / value tested) × £120,000. If this projection exceeds tolerable misstatement, the auditor should extend procedures. Otherwise, if within limits, no further action is needed.

Evaluating Results and Responding to Deviations

The auditor evaluates actual errors and projects them to the entire population. If projected errors exceed acceptable limits, further procedures or expansion of the sample may be needed. All findings and decisions must be documented.

If significant, deviations or misstatements must be reported to management and those charged with governance.

Worked Example 1.3

Scenario:
A sample of 100 payroll transactions uncovers 6 unauthorised payments. The auditor set a tolerable deviation rate of 4%.

Question:
What should the auditor do?

Answer:
The actual deviation rate (6%) exceeds the tolerable limit (4%). The auditor cannot rely on controls and should increase substantive testing in this area.

Exam Warning

Sampling is not suitable where the population is very small or highly variable; in these cases, test all items. Never use block selection as your only method unless clearly required—this could result in bias and loss of marks.

Documentation Requirements (ISA 530)

Proper documentation is essential for audit quality and review. According to ISA 530, auditors must document:

• The sampling method and rationale.
• Population details and sample size.
• Selection procedures.
• Nature, timing, and extent of work performed on each item.
• Results, including deviations/misstatements found.
• The auditor’s evaluation and conclusions.

Revision Tip

Always link your sampling documentation to the overall audit strategy and risk assessment to demonstrate professional judgement.

Comparison: Statistical vs Non-Statistical Sampling

Key Point Checklist

This article has covered the following key knowledge points:

• The definition and purpose of audit sampling.
• Differences between statistical and non-statistical sampling.
• Various selection methods (random, systematic, haphazard, block, monetary unit).
• The meaning of sampling and non-sampling risk.
• How to design, perform, and evaluate a sample as required by ISA 530.
• Documentation standards for sampling.
• The importance of stratification, tolerable misstatement, and projecting errors.

Key Terms and Concepts

• audit sampling
• sampling risk
• non-sampling risk
• statistical sampling
• non-statistical sampling
• tolerable misstatement
• stratification