AUD 689 Audit Sampling Notes
Audit Sampling Overview
- Audit sampling involves applying audit procedures to less than 100% of items, ensuring each sampling unit has an equal chance of selection (ISA 530).
- It can be statistical or non-statistical.
Audit Evidence (ISA 500)
- Auditors need sufficient, appropriate audit evidence to form audit opinions, often using sampling.
Sampling Risk
- Sampling risk is the risk that the auditor's conclusion based on a sample differs from the conclusion from a 100% examination.
- Auditors can lower sampling risk by increasing the sample size.
Types of Sampling Risk
- Compliance/Control Testing:
- Type 1: Assessing control risk too high (incorrectly concluding high risk).
- Type 2: Assessing control risk too low (incorrectly concluding low risk).
- Substantive Testing:
- Type 1: Risk of concluding material error exists when it doesn't.
- Type 2: Risk of concluding material error doesn't exist when it does.
Non-Sampling Risk
- Non-sampling risk arises from factors other than sample size (e.g., misinterpreting evidence).
- Minimized through proper planning, supervision, and quality control.
Types of Audit Sampling
- Statistical (Probabilistic) Sampling
- Non-Statistical (Non-Probabilistic or Judgmental) Sampling
Statistical Sampling
- Involves random sample selection and probability theory to evaluate results and measure sampling risk.
- Allows reliance on smaller samples and systematic approach
Non-Statistical Sampling
- Relies on auditor's professional judgment for sample selection and evaluation.
- More subjectivity in sampling decisions.
Categories of Statistical Sampling
- Attribute Sampling
- Monetary Unit Sampling
- Classical Variable Sampling
Attribute Sampling
- Estimates the proportion of a population with a specific characteristic (e.g., deviation rate in controls).
- Example: Credit check performed or not
Monetary Unit Sampling
- Uses attribute sampling to estimate misstatement amount.
- Suited for detecting overstatement errors, especially with low error rates.
- Expresses conclusions in monetary amounts.
- E.g., auditing accounts receivable, loans, investment securities and inventory
Classical Variable Sampling
- Uses normal distribution theory.
- Estimates amount for transactions or account balances and determine material misstatements.
- E.g., auditing account receivable in which a large amount of misstatement is expected, and inventory
When to Use Statistical Sampling
- When accounting/internal control systems are reliable/effective
- Large/Homogeneous populations
- Identifiable/Accessible items
Advantages of Statistical Sampling
- Efficient sample design
- Measures sufficiency of evidence
- Quantifies sampling risk
- Smaller Sample Size
- Systematic and Scientific Approach
Disadvantages of Statistical Sampling
- Additional costs include auditor training and software
Non-Statistical Sampling
- Less costly and time-consuming.
Advantages of Non-Statistical Sampling
- Lower training costs, ease of implementation, proposed adjustment based on quantitative analysis
Disadvantages of Non-Statistical Sampling
- Training difficulties, absence of consistency and uniformity, potential litigation, may cause incorrect evaluation of sampling risk
Similarities Between Statistical and Non-Statistical Sampling
- Sample design, sample selection, performing audit procedures, and evaluating results.
Differences Between Statistical and Non-Statistical Sampling
- Statistical sampling relies on statistical formulas while non-statistical relies on experience.
- Statistical sampling is objective; non-statistical is subjective.
- Different auditors may have different non-statistical approaches.
- Statistical sampling can delegate the responsibility to lower rank staff while non-statistical need senior audit staff.
- Statistical sampling is defensible in court of law
Application of Sampling on Audit Tests
- Planning the sample and design the sample size
- Selecting the sample
- Testing the sample (Vouching)
- Evaluating the results
1) Planning the Sample
- Consider audit procedure objectives and population attributes.
- Population is the entire data set from which the sample is selected.
Steps in Planning the Sample
- Determine test objectives.
- Define errors or deviations sought.
- Identify population and sampling unit.
- Stratify population.
- Decide sample size.
Stratification of Population
- Dividing a population into subpopulations with similar characteristics such as Monetary value
Factors Influencing Sample Size
- Test of Control: Auditor's intended reliance on internal control, tolerable error rate, expected error rate, and confident level needed.
- Substantive Procedures: Assessment of inherent and control risk, use of other substantive procedures, confident level needed, tolerable error, and expected error rate.
2) Selecting the Sample
- Samples should represent the population, with all items having an equal chance of selection.
- Statistical requires random selection; non-statistical uses auditor judgment.
Principal Methods of Selecting Samples
- Random Sampling
- Systematic Selection
- Haphazard Selection
Random Sampling
- Each item has an equal chance of selection; computer or numbers tables used.
- Advantage: No auditor bias.
Systematic Selection
- Sampling interval determined by dividing population by sample size. A starting number is selected in the first interval and then every nth item is selected
Haphazard Sample
- Selection without specific reason or structured technique; avoid bias.
- Useful for non-statistical sampling, but auditor tend to bias.
3) Testing the Sample
- Perform appropriate audit procedures on each item selected.
- Replace inappropriate samples (e.g., voided transactions).
- Perform alternative tests if unable to apply planned procedures (e.g., missing invoices).
4) Evaluating the Results
- Examine implications of each error discovered by analyzing its nature and cause.
- Consider qualitative aspects such as disclosure requirements and potential fraud.
- Project error to population and compare with acceptable error.
Draw Final Conclusion
- Evaluate sample result and determine whether assessment of population is confirmed or needs to be revised.
- If errors indicate ineffective control or material misstatement:
- Request management investigate and adjust.
- Modify further audit procedures.
- Consider effect on the audit report.