Audit sampling allows auditors to form conclusions about entire populations by testing representative subsets. It's a practical necessity — testing every transaction in large organizations would be prohibitively expensive and time-consuming.
Types of Audit Sampling
Statistical Sampling
- Uses probability theory to select samples and evaluate results
- Allows quantification of sampling risk
- Methods include:
- Random sampling: Every item has an equal chance of selection
- Systematic sampling: Select every nth item
- Stratified sampling: Divide population into subgroups, sample each
- Monetary unit sampling (MUS): Probability proportional to size
Non-Statistical Sampling
- Relies on auditor judgment rather than statistical methods
- Selection methods include:
- Haphazard selection: Auditor selects without bias
- Block selection: All items in a selected period
- Judgmental selection: Target specific characteristics
- Valid when properly applied, but cannot quantify sampling risk
Sample Size Determinants
Sample size is influenced by:
| Factor | Higher Sample Size When... |
|---|---|
| Risk of material misstatement | Higher risk assessed |
| Tolerable misstatement | Lower tolerance |
| Expected misstatement | Higher expected errors |
| Population size | Larger (minimal impact above certain thresholds) |
| Desired confidence | Higher confidence needed |
Sampling Process
1. Define the Objective
- What assertion is being tested?
- What constitutes a deviation or misstatement?
2. Define the Population
- Ensure completeness of the population
- Define the sampling unit (invoice, transaction, dollar)
3. Determine Sample Size
- Apply sampling methodology
- Consider risk factors and materiality
4. Select the Sample
- Use appropriate selection method
- Document selection criteria
5. Perform Procedures
- Apply audit procedures to selected items
- Document results for each item
6. Evaluate Results
- Project sample results to the population
- Consider qualitative aspects of errors found
- Determine impact on audit opinion
Common Pitfalls
- Population not complete: Testing from an incomplete list
- Sample not representative: Selection bias
- Inappropriate projection: Extrapolating incorrectly
- Ignoring qualitative factors: Focusing only on dollar amounts
- Under-sampling: Too few items to support conclusions