Corporate Accounting and Auditing · Audit Sampling, Audit Techniques and Analytical Procedure

Audit Sample Design, Size and Evaluation of Results

Updated 10 October 2026 · Fact-checked

Under SA 530, you design a sample by considering the purpose of the procedure and the population, choose a size large enough to reduce sampling risk to an acceptably low level, select items so every unit has a chance of selection, project misstatements found in tests of details, and evaluate whether the sample gives a reasonable basis for conclusions.

Understand Sample Design, Size and Evaluation of Results

Audit sampling means applying audit procedures to less than 100% of the items in a population, so that every sampling unit has a chance of being selected. The auditor then uses what the sample shows to form or support a conclusion about the whole population. SA 530 allows both statistical and non-statistical approaches.

The first step is sample design. SA 530 (para 6) says you must consider the purpose of the audit procedure and the characteristics of the population. You need a clear idea of what counts as a deviation (in tests of controls) or a misstatement (in tests of details). Only conditions relevant to the purpose of the procedure are counted. Example from the standard: in a confirmation test of existence of receivables, a customer payment made before the confirmation date but received shortly after it is not a misstatement. A misposting between customer accounts does not change total receivables, so it may not count either, although it may matter for fraud risk or the allowance for doubtful accounts.

Next is sample size. Para 7 requires a size sufficient to reduce sampling risk to an acceptably low level. Size can come from a statistically based formula or from professional judgment. When circumstances are similar, the factors affect size in the same way whichever approach you use. The expected error also matters. In tests of controls, you assess the expected rate of deviation. If it is unacceptably high, you will normally decide not to perform tests of controls. In tests of details, you assess the expected misstatement. If it is high, 100% examination or a large sample may be appropriate.

After testing, you project misstatements. For tests of details, para 14 requires you to project the misstatements found in the sample to the population. This gives a broad view of the scale of misstatement. As A18 notes, the projection may not be enough to decide the amount to be recorded.

Finally, you evaluate results (para 15): the results of the sample, and whether sampling has given a reasonable basis for conclusions about the population tested. If it has not, you may ask management to investigate the misstatements and the potential for further ones and make adjustments, or tailor the nature, timing and extent of further procedures. For tests of controls, you might extend the sample, test an alternative control or modify related substantive procedures.

Key rules to remember

Simple projection of misstatement (ratio method): Projected misstatement = (Misstatement found in sample ÷ Value of sample) × Value of population; A common way to apply the para 14 requirement when the misstatement is proportional to value. Use it for the population from which the sample was drawn.
Projection by average error (difference method): Projected misstatement = (Misstatement found ÷ Number of items in sample) × Number of items in population; Useful when errors do not depend on the size of the item. State clearly which method you used.
Sample deviation rate (tests of controls): Sample deviation rate = Number of deviations found ÷ Sample size; Compare with the tolerable rate you set. The sample rate is the auditor's best estimate for the population rate.
SA 530 requirement on size: Sample size must reduce sampling risk to an acceptably low level; Para 7. Size comes from a statistical formula or professional judgment.
SA 530 requirement on selection: Every sampling unit in the population must have a chance of selection; Para 8.

How to solve Sample Design, Size and Evaluation of Results questions

Use this order for any SA 530 question on design, size or evaluation. It keeps your answer tied to the standard.

1Identify the purpose of the procedure: a test of controls (deviations) or a test of details (misstatements).
2Define the population and what counts as a deviation or misstatement. Exclude conditions not relevant to the purpose.
3State the factors for sample size: the expected deviation or misstatement, the sampling risk you will accept, the tolerable level, and the population characteristics. Say whether size rises or falls.
4Select items so each sampling unit has a chance of selection. Name the method if the question asks.
5Do the calculation: count deviations, or project misstatements to the population using the stated method.
6Compare the result with the tolerable level and judge whether it gives a reasonable basis for a conclusion.
7If it does not, state the action: ask management to investigate and adjust, extend the sample, test an alternative control, or change the substantive procedures.
8 Write a clear conclusion in one line.

Quickest way: Purpose, size, select, project, conclude

When to use it: For short theory questions and for numerical questions that ask you to project misstatements and judge the result.

Write the purpose and the definition of a misstatement or deviation.
List size factors in one line each, with the direction of effect.
For numbers, compute the rate or projected figure first, then compare it with the tolerable level.
If the projection exceeds tolerable level or the sample is inconclusive, name two follow-up actions from SA 530 A23.
End with one sentence stating that the projection gives a broad view and may not fix the amount to be booked.

Common mistakes in Sample Design, Size and Evaluation of Results

Counting every difference found as a misstatement.
Students forget that only conditions relevant to the purpose of the procedure count.
Fix: Define the misstatement first. In a receivables existence test, a payment received just after the confirmation date is not a misstatement.
Saying the projected misstatement is the amount the client must book.
The projection looks like a precise figure.
Fix: State that projection gives a broad view of the scale of misstatement and may not be sufficient to determine an amount to be recorded.
Saying a higher expected error allows a smaller sample.
Confusion between tolerable and expected error.
Fix: If expected misstatement is high, a larger sample or 100% examination may be appropriate. A higher tolerable level, by contrast, allows a smaller sample.
Projecting the sample error to the sample value instead of the population.
Rushing through the ratio.
Fix: Divide by the sample value and multiply by the population value. Check that the answer is larger than the error found.
Ignoring what to do when the sample is inconclusive.
Students stop at the calculation.
Fix: Add the para A23 actions: management investigation and adjustment, or change in nature, timing and extent of procedures, such as extending the sample.
Claiming that only random selection satisfies SA 530.
Mixing up statistical sampling with the general requirement.
Fix: The standard requires that every sampling unit has a chance of selection. It allows statistical and non-statistical approaches.

Worked examples

Example 1

While auditing trade receivables of Kaveri Textiles Ltd, a CMA selects a sample of items worth ₹8,00,000 from a population of ₹60,00,000. Overstatements of ₹16,000 are found in the sample, all proportional to value. Project the misstatement and state what the auditor should do next if the tolerable misstatement is ₹1,00,000.

Show the solution

Purpose: test of details on existence and valuation, so SA 530 para 14 requires projection.
Misstatement rate = ₹16,000 ÷ ₹8,00,000 = 2%.
Projected misstatement = 2% × ₹60,00,000 = ₹1,20,000.
Compare: ₹1,20,000 exceeds tolerable misstatement of ₹1,00,000.
So the sample has not given a reasonable basis for concluding that receivables are free from material misstatement.
Under A23, the auditor may ask management to investigate the misstatements and the potential for further ones and adjust, or extend the sample or modify further procedures.
The projection gives a broad view of scale and may not be enough to decide the amount to be recorded.

Answer: Projected misstatement is ₹1,20,000, above the tolerable ₹1,00,000. The auditor should ask management to investigate and adjust, or tailor the nature, timing and extent of further procedures, for example by extending the sample.

Example 2

Explain how the auditor should treat the following in a confirmation test of accounts receivable existence: (a) a customer payment made before the confirmation date but received by the client shortly after it, and (b) a misposting between two customer accounts. Also state two factors that increase sample size.

Show the solution

Under SA 530 A6, the auditor must understand what is a misstatement for the purpose of the procedure, so only relevant conditions are included in the projection.
(a) The payment was made before the confirmation date, so the balance confirmed by the customer differs only because of timing. It is not a misstatement for this test.
(b) A misposting between customer accounts does not change total receivables. It may not be a misstatement for this test.
However, (b) may matter for other areas, such as risk of fraud assessment or the adequacy of the allowance for doubtful accounts, so the auditor should still consider it there.
Factors increasing size: a high expected misstatement in the population (para A7 says a large sample or 100% examination may be appropriate), and a lower level of sampling risk the auditor is willing to accept.
Whatever the size, para 7 requires that it reduces sampling risk to an acceptably low level.

Answer: Neither item is normally counted as a misstatement in this existence test, but the misposting may affect fraud risk and the doubtful debt allowance. Higher expected misstatement and a lower acceptable sampling risk both increase sample size.

Exam tips

In theory questions, quote the SA 530 requirements in short form: para 6 design, para 7 size, para 8 selection, para 14 projection, para 15 evaluation.
For numerical questions, show the rate, the projection and the comparison with tolerable misstatement in separate lines to earn step marks.
Always add the follow-up action when the result is unfavourable. Examiners often award a mark for it.
In MCQs, watch for options that say projection fixes the amount to be booked. That is wrong, since SA 530 says it may not be sufficient.
Do not give a paragraph number you are unsure of. State the rule in words instead.

Practice questions from Audit Sampling, Audit Techniques and Analytical Procedure

Sample Design, Size and Evaluation of Results in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

ACCA Applied SkillsDesigning Samples and Evaluating Results

Sample Design, Size and Evaluation of Results: frequently asked questions

What factors affect audit sample size?

SA 530 requires a size sufficient to reduce sampling risk to an acceptably low level. The expected deviation rate in tests of controls and the expected misstatement in tests of details matter. If the expected misstatement is high, a large sample or 100% examination may be appropriate.

Why must the auditor project misstatements?

For tests of details, SA 530 para 14 requires projection of misstatements found in the sample to the population. This shows the likely scale of misstatement in the whole population. It may not be enough to decide the amount to be recorded.

What if the sample does not give a reasonable basis for a conclusion?

The auditor may ask management to investigate the misstatements and the potential for further ones and make adjustments. Or the auditor may change the nature, timing and extent of further procedures. For tests of controls, this can mean extending the sample or testing an alternative control.

Does SA 530 require statistical sampling?

No. Audit sampling can use statistical or non-statistical approaches. Size can be set by a statistical formula or by professional judgment. In both cases, every sampling unit must have a chance of selection.