# SA 530 – Audit Sampling
## What is Audit Sampling?
Audit Sampling = Application of audit procedures to less than 100% of items within a population of audit relevance, such that all sampling units have a chance of selection to provide a reasonable basis for conclusions about the entire population.
Key insight: The extent of checking is a matter of auditor's judgment. Nothing statutorily mandates how much work must be done. It is not obligatory for the auditor to adopt a sampling technique — the auditor is bound only by his opinion on the FS.
## Key Definitions
| Term | Meaning |
|---|---|
| Sampling Unit | The individual items that make up the population. |
| Population | Entire set of data from which a sample is selected & about which auditor wishes to draw conclusions. |
| Anomaly | Misstatement/deviation that is NOT representative of misstatements/deviations in the population. |
| Tolerable Misstatement | Monetary amount set by the auditor for which he seeks reasonable assurance that the actual misstatement in the population does not exceed this amount. |
| Tolerable Rate of Deviation | Rate of deviation from internal control set by auditor, for which he seeks assurance that actual rate of deviation does not exceed it. |
## Sampling is Performed On:
- Tests of Controls (TOC) – to identify deviations from internal controls.
- Tests of Details (TOD) – to identify misstatements in Classes of Account/Disclosures.
## Characteristics of a Good Population
1. Appropriateness – Population must be relevant to the specific audit objective.
2. Completeness – Must include all activities forming part of the relevant control, throughout the entire period (for TOC).
3. Reliability – Accuracy and source must be reliable; else samples drawn lack relevance.
Sample must be representative – Whether statistical or non-statistical, the sample must closely resemble the whole population (not necessarily identical).
## Approaches to Sampling (Types)
### A. Statistical Sampling
- Uses random selection of sampling units AND probability theory to evaluate sample results, including measurement of sampling risk.
- Recommended for large organisations with huge transaction volumes — unbiased, samples not prejudged.
Features:
- Audit testing is more scientific (uses mathematical laws of probability).
- Wide application where population consists of large number of similar items.
- No personal bias of auditor → more reliable.
Advantages of Statistical Sampling:
- Sample size does NOT increase in direct proportion to population size.
- Sample selection is more objective and defensible.
- Provides means of estimating minimum sample size related to specified risk.
- Provides means of deriving a 'calculated risk' and corresponding sample size.
- Provides better description of large data than a complete examination.
- Widely accepted, scientific, without personal bias.
### B. Non-Statistical Sampling
Sampling based on personal experience and knowledge of the auditor.
## Sampling Risk vs. Non-Sampling Risk
### Sampling Risk
The risk that the auditor's conclusion based on a sample may differ from the conclusion if the entire population were tested.
Leads to two types of erroneous conclusions:
| Type | TOC | TOD | Effect |
|---|---|---|---|
| Type 1 (Worse) | Controls more effective than they actually are | MM does not exist when it actually does | Affects audit effectiveness → likely inappropriate opinion |
| Type 2 (Better) | Controls less effective than they actually are | MM exists when it actually does not | Affects audit efficiency → additional unnecessary work |
### Non-Sampling Risk
The risk that the auditor reaches an erroneous conclusion for any reason NOT related to sampling (e.g., inappropriate audit procedures, misinterpretation of evidence).
## Factors Deciding Extent of Checking (Sampling Plan)
- Size of the organisation.
- State of Internal Control.
- Adequacy and reliability of books and records.
- Degree of desired confidence.
- Tolerable error range.
## Precautions while Applying Test Check Techniques
1. Thorough study of the accounting system before sampling.
2. Proper study of Internal Control System (ICS).
3. Areas unsuitable for sampling should be carefully considered.
4. Proper planning of sampling methods and explanation to staff.
5. Transactions/balances must be properly classified (stratified).
6. Sample size appropriately determined.
7. Sample chosen in an unbiased way.
8. Errors located in the sample analysed properly.
## Sampling Process
Sample Design → Sample Size → Sample Selection → Audit Procedures → Nature & Cause of Deviation → Projecting → Evaluating Results
### Sample Design
When designing, the auditor considers:
- Audit procedures to achieve specific purpose and best combination.
- Nature of evidence required and possible deviation/misstatement.
- Procedures to obtain evidence about completeness of population.
### Stratification
- Dividing a heterogeneous (diversified) population into homogeneous sub-populations.
- Each sub-population = stratum; units within = strata.
- Samples are drawn from each sub-population.
- Improves audit efficiency & reduces variability within each stratum → reduces sample size without increasing sampling risk.
- The auditor must combine results across strata for overall conclusion.
### Value-Weighted Selection
- Identifies sampling unit as individual monetary units that make up the population.
- Audit effort is directed to larger value items (greater chance of selection).
- Can result in smaller sample sizes.
### Sample Size
Auditor shall determine sample size sufficient to reduce sampling risk to an acceptably low level. The lower the risk the auditor is willing to accept, the greater the sample size needs to be.
Factors Influencing Sample Size:
| Factor | Effect on Sample Size |
|---|---|
| Greater reliance on operating effectiveness of controls | Increases (larger TOC) |
| Higher assessment of ROMM | Increases |
| Stratification applied | Decreases |
| Increase in tolerable rate of deviation / tolerable misstatement | Decreases |
| Higher expected rate of deviation/misstatement | Increases |
| Increase in desired level of assurance | Increases |
| Increase in number of sampling units in population | Negligible effect |
### Sample Selection Methods
1. Random Sampling – All items have a known chance of selection.
- Simple Random Sampling – Each unit has equal chance; suitable for homogeneous populations.
- Stratified Sampling – Heterogeneous population divided into homogeneous sub-populations; sample drawn from each stratum.
2. Interval (Systematic) Sampling – Number of units in population ÷ sample size = sampling interval (e.g., 50). Determine a starting point within the first 50; select every 50th unit thereafter. Risk: sampling interval should NOT correspond to a pattern in the population. Use >1 starting points to minimise risk.
3. Monetary Unit Sampling – A type of value-weighted selection.
4. Haphazard Sampling – Auditor selects without a structured technique. NOT appropriate in statistical sampling.
5. Block Sampling – Selection of block(s) of contiguous (adjacent) items. Generally cannot be used as items in sequence may have similar characteristics, different from items elsewhere in population.
### Nature and Cause of Deviations/Misstatements
- Auditor analyses if deviations/misstatements share a common feature — if so, identify all items in population with that feature and extend procedures.
- Investigate nature and cause and evaluate possible effect.
- Anomaly classification requires the auditor to obtain a high degree of certainty that the deviation is NOT representative of the population. (Rare circumstance.)
### Projecting Misstatements
- Auditor must project misstatements to obtain a broad view of misstatement scale.
- An anomaly is excluded when projecting (but the uncorrected anomaly itself is still considered alongside the projection).
- For TOD → auditor projects misstatements from sample to population.
- For TOC → no explicit projection of deviations needed (sample deviation rate IS the projected deviation rate for the population).
### Evaluating Results of Sampling
The auditor evaluates:
- Results of the sampling.
- Whether use of sampling has provided reasonable basis for conclusion about the population.
## Quick Memory Aid
Population characteristics (ACR): Appropriate, Complete, Reliable.
Sample selection methods: Random, Interval, Monetary, Haphazard, Block (RIMHB).