Audit Evidence, Sampling, and Analytical Procedures
This domain tests whether evidence is sufficient, relevant, and reliable, whether sampling supports extrapolation, and whether analytical procedures are properly designed.
Evidence
- Sufficiency is quantity; relevance is the logical link to the criterion; reliability depends on source and nature.
- Procedures include inspection, observation, inquiry, confirmation, recalculation, re-performance, and analytical procedures.
- Auditor’s direct evidence and external confirmations are generally stronger than internal records; originals are stronger than copies.
Sampling
- Statistical or non-statistical; random, systematic, or monetary unit sampling.
- Define population, sampling unit, sample size, and selection method before testing.
- Project errors only from a representative sample; investigate anomalies separately.
Analytical procedures
- Trend, ratio, reasonableness, and regression analysis; set an expectation and a threshold for investigation.
Worked example
Payroll population: 1,200 employees, 480 million tenge. Sample: 60 employees, value 24 million; three overpayments total 0.12 million. Error rate = 5% (0.12/24). Projected error = 5% × 480 million = 24 million. The auditor reports the projection, investigates the cause, and does not report 0.12 million as the total likely error. If selection was biased, projection is invalid.
Traps
- Relying on inquiry alone for a significant assertion.
- Extrapolating from a haphazard or non-representative sample.
- Ignoring an outlier that suggests fraud.