## SA 530: Sample Design, Size & Selection Methods

### Key Definitions

Non-Statistical Sampling — A sampling approach that lacks either (i) random selection or (ii) probability theory to evaluate results.

Stratification — Dividing a population into sub-populations (strata), each containing sampling units with similar characteristics (often monetary value), before sampling each stratum separately.

---

### Five Requirements When Using Audit Sampling

1. Sample Design, Size & Selection of Items

2. Performing Audit Procedures

3. Nature and Cause of Deviations and Misstatements

4. Projecting Misstatements

5. Evaluating Results of Audit Sampling

---

### (A) Sample Design

Consider:

Purpose of the audit procedure
Characteristics of the population

### (B) Sample Size

Large enough to reduce sampling risk to an acceptably low level.

### (C) Selection of Items

Every sampling unit in the population must have a chance of selection (representative sample).

---

### Methods of Selecting Samples

Method	How It Works	Key Rule
Random Selection	Random number generators or tables	Every unit has equal, calculable probability
Systematic Selection	Population ÷ Sample size = interval; pick every nth item from a random start	Start point should be randomised
Monetary Unit Sampling (MUS)	Value-weighted; each monetary unit is a sampling unit	Conclusion expressed in monetary amounts
Haphazard Selection	No structured technique; auditor avoids conscious bias	NOT valid for statistical sampling

### Systematic Selection — Step by Step

1. Interval = Population size ÷ Sample size (e.g., 500 ÷ 10 = 50)

2. Randomly choose a start within the first interval (e.g., item #17)

3. Select: 17, 67, 117, 167 … until the population is exhausted

### Haphazard vs Random — Key Distinction

	Haphazard	Random
Formal mechanism?	No	Yes (tables/software)
Conscious bias eliminated?	Yes (by intent)	Yes (by design)
Valid for statistical sampling?	No	Yes

Worked example

### Example 1

Systematic Selection: Population = 2,000 purchase invoices; required sample = 40. Interval = 2,000 ÷ 40 = 50. Random start (from table) = 23. Sample items: 23, 73, 123, 173 … 1,973.

### Example 2

MUS (Monetary Unit Sampling): Debtors ledger total = ₹10,00,000; desired sample = 10 monetary units of ₹1,00,000 each. A debtor with a balance of ₹5,00,000 has a 5-in-10 chance of selection, giving high-value accounts proportionally greater coverage. The final conclusion is stated as: 'Misstatement in debtors does not exceed ₹X.'

### Example 3

Stratification before sampling: A payables population of ₹50 lakh is split into: (i) items > ₹1 lakh — audited 100%; (ii) items ₹10,000–₹1 lakh — sampled at 20%; (iii) items < ₹10,000 — sampled at 5%. This reduces sampling risk on high-value items without over-sampling trivial ones.

⚠️ Common exam mistakes

Using haphazard selection for statistical sampling — haphazard selection is only valid for non-statistical sampling.
Confusing systematic selection with random selection — systematic uses a fixed interval from a random start, not a fresh random draw for every item.
Forgetting that MUS provides a conclusion in monetary amounts, not merely a count of errors or deviations.
Treating stratification as a sampling method — it is a population-division technique applied before selecting a sample, not a selection method itself.

Bare-Act text Para 11 · SA 530 - Audit Sampling (ICAI) · click to expand

The auditor shall select items for the sample in such a way that each sampling unit in the population has a chance of selection.

Now that you've read this — what's next?

Move from understanding → mastery in 3 clicks. Each option below picks up from this lesson's topic.

🔥 Drill 5 Qs on this → ✍️ Adaptive practice →

Sample Design, Size, and Methods of Selecting Items

Worked example

⚠️ Common exam mistakes