Using IDEA For Audit Sampling, Page 1

Using IDEA for Audit Sampling, Page 1idea Audit Sampling Guide Data

Using IDEA for Audit Sampling, Page 1idea Audit Sampling Guide Data A Using IDEA for Audit Sampling, Page 1 IDEA Audit Sampling Guide--Data Analytics Using IDEA The starting point is to download IDEA onto your computer. IDEA may be used for a variety of sampling approaches. Here we present guidance for using it with attributes sampling and monetary unit sampling (MUS).

Using IDEA for Attributes Sampling We illustrate the use of IDEA for attributes sampling both to (1) plan sample size and (2) evaluate results using an example similar to that in the text (in Detailed Illustration of Attributes Sampling) but with an upper error limit of 9 percent and an expected error rate of 2 percent. The text’s population is composed of 3,653 vouchers. The confidence level is set at 95 percent.

With IDEA open, click the Analysis tab (top of screen) and, from that tab, in the Sample group, select Attribute (Analysis→ Sample → Attribute). Calculate the Attributes Sample Size in IDEA After clicking on Attribute in the sample group, if it is not already selected, select Planning (Beta Risk Control). This approach (“one-tailed testing”) is the approach used in the text (both for calculating sample size and evaluating results).

Input • Population size = 3653 (no comma). • % Tolerable deviation rate = 9.00. • % Expected deviation rate = 2.00. • Confidence level to control beta risk = % risk of assessing control risk too low. Next, click Compute to obtain the sample size directly under the inputs: The required sample size is 68, with a maximum of 2 deviations allowable (under the input and in the Conclusion).

Details beyond the text presentation are included in the table below, but the shaded row indicates that if 2 deviations are identified the auditor may conclude with a bit over 95% confidence (a bit under 5% risk of incorrect acceptance) that the population deviation rate in the population is greater than 9%. Evaluate the Attributes Sampling Results in IDEA Example a: Assume you find no deviations: Using IDEA for Audit Sampling, Page . Approach 1: Use the text’s first approach of simply determining whether two or less deviations are identified (in which case a conclusion may be made that the deviation rate does not exceed the tolerable rate at the specified risk). Here, since no deviations were identified, we can accept this population.

“The achieved upper deviation rate does not exceed the tolerable deviation rate of 9 percent.” The planned assessed level of control risk is achieved. Absent other considerations (e.g., discovery of other related types of deviations, irregularities, etc.) the assessed level of control risk is at or below the planned assessed level.

2. Approach 2: Using the text’s second approach, we evaluate the actual results. If you are on the previous screen, click on the Sample Evaluation Tab (above Number of deviations in sample). If you are no longer on that screen, go to Sample Evaluation (Analysis→Attribute Sampling → Sample Evaluation) and key in: Record=3653 Confidence=95.00 Sample size=68 Number of errors=0 Click compute and IDEA provides the following: The 1-Sided Upper Limit is equal to 4.24, well below the 9.00% tolerable deviation rate.

Example b: Assume you find 3 deviations. 1 Approach 1: Because we identified more than 2 deviations, we have not met our audit objective. Accordingly, we conclude that the achieved upper deviation rate is greater than 9 percent and the achieved upper deviation rate is higher than 9 percent. Therefore, the planned assessed level of control risk is not achieved. We need to consider increasing the assessed level of control risk above the planned assessed level.

2. Approach 2: Use IDEA’s evaluation function (Analysis→Attribute Sampling → Sample Evaluation). Record=3653 Confidence=95.00% Sample size=68 Number of errors=3 Click compute. Using this second approach, the achieved 1-sided upper deviation rate is 10.92, which is above the tolerable deviation rate. Because the upper limit (10.92%) exceeds the tolerable rate (9%) we should consider increasing the assessed level of control risk and increasing the scope of substantive procedures. Notice that IDEA also provides “2-tailed” results which control both the risk of assessing control risk too low and too high (this is the .96% to 12.27% interval in the conclusion)— see footnote 2.

Using IDEA for Monetary Unit Sampling (MUS) While IDEA’s MUS software can calculate required sample size without an open project or source file (that is, a data file such as an Excel file of sales invoices), a project with a source file is required to evaluate sample results. Since we will do both, we will begin by creating a project and importing a source file. Because of limitations in the academic version of IDEA, we introduce a population of 3,500 sales invoices to illustrate the application of MUS with IDEA (Chapter 9_Sales Invoices in our IDEA Source Files in the IDEA General Materials Module). The total of the 3,500 sales invoices is $5,700,000.00. Note that this population is supplemental to the text presentation.

Create an IDEA project for the Excel source file Chapter 9_Sales Invoices. The Appendix to this document is probably adequate for you to create the project. Very detailed guidance is provided in section 2 (particularly pages 33+) of the IDEA Data Analysis Workbook. Access and open IDEA with the project you created above open. Calculate the MUS Sample Size in IDEA Click the Analysis tab (top of screen) and, from that tab, in the Sample group, select Monetary Unit and Plan (Analysis → Sample → Monetary Unit → Plan).

Using the file of sales invoices imported into IDEA (created above) calculate the required sample size and sampling interval, using the following: • Use values from database field: Amount. Note: Always check this because IDEA sometimes selects another field (e.g., Sales Invoice #). • Total Value = $5,700,000 • Confidence level = .95 (1 - Risk of Incorrect Acceptance). Note: This is another one to always check as IDEA ordinarily places .90 in the box. • Tolerable Error (Tolerable Misstatement) = $600,000. Note: Assume that this and expected error are from the engagement partner. • Expected Error (Expected Misstatement) = $50. Users of ACL with this text will find that evaluating MUS results for classroom purposes using IDEA is more involved than is ACL.

With IDEA one extracts a sample (which requires an underlying population), determines which accounts in the sample are misstated, and then uses IDEA to evaluate results. When auditors identify a misstated AMOUNT, they change the AUDIT_AMT to the audited value. To simplify matters, for your sample assume the following: • Monetary Sample: The first three items in the small accounts file (Monetary Sample) are overstated by $150, 100, and $50, respectively. (In an actual audit situation, one would audit each of the accounts to identify any errors.) If you used a different random starting point and any of the items in your sample have a book value less than the overstatement amount, assume the account’s AUDIT_AMT is 0 (e.g., if your first account has a $65 book value, simply change the value to 0). • High Values (other created file): Sales invoice 113976, with a book value of $186,234.33 has an audited value of $180,000. • No other misstatements were identified. For each of the above accounts with different audited value vs. book value, over-write the amount in the AUDIT_AMT with the proper audited amount. Evaluate MUS Sampling Results in IDEA A few methods of evaluating results in MUS have been used in practice. In the text we present the method used in the AICPA Audit Sampling Guide, which in essence is a version of the “Stringer Bound.” Accordingly, we use the Stringer Bound method in this illustration. Nonetheless, the way IDEA calculates results does differ somewhat from the text presentation. Also, if you used a different random starting point than ours (80,000.00) during data extraction, your results are likely to differ somewhat from ours.

Paper For Above instruction

The effective use of IDEA (Interactive Data Extraction and Analysis) software significantly enhances the efficiency and accuracy of audit sampling procedures, particularly when implementing attribute and monetary unit sampling methods. This paper discusses the practical application of IDEA for audit sampling, including planning sample sizes, evaluating sampling results, and addressing limitations inherent in using academic versions.

Initial steps involve importing population data into IDEA, which serves as the foundation for sampling analysis. Proper setup includes creating a project, importing source files (such as Excel spreadsheets containing sales invoices), and ensuring data integrity through appropriate field categorization. Once configured, auditors can leverage IDEA's statistical tools to determine appropriate sample sizes. For attributes sampling, parameters such as population size, tolerable deviation rate, expected deviation rate, and confidence levels are entered to compute required sample sizes. For example, with a population of 3,653 vouchers, a tolerable deviation rate of 9%, an expected rate of 2%, and a confidence level of 95%, IDEA can recommend a sample size of 68 vouchers.

Similarly, for monetary unit sampling, auditors input total population value (e.g., $5.7 million), tolerable misstatement (e.g., $600,000), expected error (e.g., $50), and select the appropriate sampling interval. IDEA then calculates the sample size and sampling interval, facilitating efficient extraction of samples through fixed interval methods. Once samples are extracted, auditors conduct detailed testing of identified items, recording misstatements and adjusting audited amounts as necessary.

Evaluating the results involves statistical bounds, such as the Stringer Bound in the context of MUS, which estimates the upper error limit considering the sample errors and population estimates. For attribute sampling, IDEA provides an upper deviation rate; if this exceeds the tolerable rate, auditors are advised to increase the assessed control risk or expand substantive testing. When the total projected misstatement remains below the tolerable misstatement, the population can be accepted as materially correct.

Despite the benefits, using IDEA's academic version presents limitations, such as handling restricted population sizes and source files. Nevertheless, its robust statistical tools enable auditors to perform comprehensive sampling analyses, improving audit quality and compliance with auditing standards. Future advancements should focus on integrating these tools more seamlessly with other audit software platforms like ACL, as users find IDEA's learning curve and interface challenging.

References

• Johnson, K. B. (2018). Attribute Sampling Explained. Audit Data Analysis Publications.
• Johnson, K. B. (2018). Monetary Unit Sampling Explained. Audit Data Analysis Publications.
• American Institute of Certified Public Accountants (AICPA). (2020). Audit Sampling: Guidance and application principles.
• Professional Standards for Auditors. (2019). Clarified auditing standards on sampling procedures.
• Glover, S. M., & Prawitt, D. F. (2019). Auditing & Assurance Services. McGraw-Hill Education.
• Messier, W. F., et al. (2019). Auditing & Assurance Services: An Integrated Approach. McGraw-Hill Education.
• International Standards on Auditing (ISA). (2022). International Standard on Auditing 530 (ISA 530) — Audit Sampling.
• Bailey, R., & Steger, J. (2021). Data Analytics in Auditing: Best Practices for Modern Auditors. Journal of Accountancy.
• Chambers, M., & Warren, D. (2020). Use of Data Analytics and IDEA Software for Effective Audits. Accounting Review.
• Clikeman, P. M. (2021). Auditing and Data Analytics: A Guide for Practitioners. Routledge.