1. Statistical vs. Non-Statistical Sampling
Auditors rarely examine 100% of transactions. Instead, they utilize audit sampling (ISA 530) to apply procedures to less than 100% of items within a population, ensuring all sampling units have an equal chance of selection:
- Statistical Sampling: Uses random selection models and probability theory to evaluate sample results and quantify sampling risk (e.g., Monetary Unit Sampling).
- Non-Statistical (Judgmental) Sampling: Relies purely on the auditor’s professional judgment and experience to select specific items (e.g., targeting high-value or high-risk balances).
2. Sampling Risk and Non-Sampling Risk
- Sampling Risk: The risk that the auditor’s conclusion based on a sample layout varies from the conclusion they would reach if the entire population were tested.
- Type I Error (Risk of Under-reliance / Incorrect Rejection): Efficiency loss; the auditor thinks controls are bad when they are actually good, expanding unnecessary testing.
- Type II Error (Risk of Over-reliance / Incorrect Acceptance): Effectiveness failure; the auditor thinks controls are great when they are actually broken. This leads directly to audit failure and liability exposure.
- Non-Sampling Risk: The risk that the auditor reaches an erroneous conclusion for any reason unrelated to sample size (e.g., using inappropriate audit procedures or misinterpreting evidence).
3. Substantive Testing for Complex Areas: Accounting Estimates
Modern accounting requires heavy management estimates (e.g., fair value measurements, expected credit loss provisions under IFRS 9, asset impairment valuations). The auditor must review the underlying assumptions used by management, test the operational data pipeline feeding the calculation, and review subsequent events occurring up to the audit report date to verify accounting estimate reasonableness.