CPA Firm Support Services, LLC
Share this content

Clarified Auditing Standards: Audit Sampling (AU-C Section 530) – Part 2

Aug 13th 2015
CPA Firm Support Services, LLC
Share this content

The first part of this article discussed requirements of AU-C Section 530. This article will discuss planning both a nonsampling and sampling application.

Planning a Strategy Not to Sample
Because sampling applications may take more time than performing nonsampling procedures, some auditors are planning strategies not to sample on smaller engagements other than certain audits of regulated industries or governments that require some form of sampling. For some recorded populations, instead of sampling, it may be more cost-effective to audit individually significant items comprising a significant portion of the recorded population.

The dollar amount of the recorded population that is considered significant depends on the monetary amounts of the account balances, on the risks of misstatement at the assertion level, on the relative difficulty of auditing a majority of a population, and on the consideration of tolerable misstatement for the financial statements as a whole. For example, confirmation of 70 percent to 80 percent of accounts receivable balances may be necessary when risk of misstatement at the financial statement and/or assertion levels is high; 60 percent or less may be sufficient when these risks are lower.

Auditors will usually audit lesser amounts of inventories than of receivables or fixed assets because of the relative cost of obtaining evidence. In doing so, less evidence is obtained supporting the opinion on the financial statements taken as a whole, which may cause evidence requirements for other account classifications to be greater. This subjective consideration will also affect decisions about the necessary audit coverage of the other account classifications.

Consistent with this reasoning, less audit coverage of account classifications that are more difficult and costly to audit may be possible if larger amounts of less costly substantive evidence can be obtained from other account balances, such as cash, fixed assets, etc. Depending on the relative amounts of these other account balances, it may be possible to audit individually significant items comprising 10 percent to 20 percent of inventories, even if risks are high, and still not sample the remaining population. As discussed elsewhere in this small audit series, risk assessment and analytical procedures may provide significant amounts of substantive evidence for all accounts, thereby reducing the need for more extensive and costly tests of balances procedures.

Planning Nonsampling Applications
When a strategy not to sample is planned, certain sampling populations may be greater than the lower limit for individually significant items. Depending on the number and characteristics of the units in the sampling population, and on the percentage of the recorded population audited by other nonsampling procedures, it may be appropriate to select a limited number of units from the sampling population for additional nonsampling procedures similar to those performed for individually significant items.

For example, individually significant items may be selected from a recorded accounts receivable population, leaving a remaining sampling population greater that the lower limit for individually significant items. If the sampling population is comprised of account balances that have a different nature than the larger balances, such as a plumbing contractor's service receivables which are generally smaller than its unpaid contract billings, it may be appropriate to select some of the smaller accounts and apply the same nonsampling procedures used for the individually significant items.

In such situations, depending on the levels of risk at the financial statement and assertion levels, it may be appropriate to select, say, five to 15 units from the sampling population. When risk is higher, a greater number of units should be selected. Sometimes called the “handful method,” low risk may permit selecting one handful (i.e., five units). High risks may require three or more handfuls. The number of units selected should be based on the auditor's professional judgment. Each of these units would be audited 100 percent.

Planning a Sampling Application
If the most efficient approach to auditing a financial statement classification is a sampling application, forms purchased from major publishers or created by auditors guide calculations and documentation. A cost-efficient approach to planning a sampling application is listed below.

  1. Determine the objective of the test (i.e., the financial statement assertion or assertions that are being evaluated).
  2. Define the population. For example, the classification for receivables may include contractual sales of products and service sales. An auditor may need to group the balances in separate populations for sampling purposes.
  3. Identify individually significant items.
  4. Determine the method of selecting the sample. Random number (using table or software), systematic (every nth unit), or haphazard (judgmental) selection methods are acceptable. Use the most efficient method for the circumstances. If haphazard sampling is used, the audit documentation must include a description of how the auditor avoided bias in selecting sample units.
  5. Determine if stratification of the population is necessary. Nonstatistical methods for sampling that are based on probability-proportionate-to-size (PPS) statistical sampling methods, like the model approach illustrated in the auditing standards, often suggest two-thirds of the sampling population for the upper stratum; one-third for the lower. A common alternative to stratification that may be more cost-efficient is to add 20 percent to the calculated sample size when there are significant variations in the size of sample units.
  6. Calculate the sample size by dividing the amount of the sampling population by tolerable misstatement for that financial statement classification. Multiply the result by a factor that reflects risks on the engagements. From the model approach to sampling in the auditing standards, this factor can range from 1.0 to 3.0 depending on low or high risks of misstatement and other procedures risk. Other procedures risk is higher when no or few other procedures are being performed to evaluate applicable assertions.
  • Perform the auditing procedures.
  • Evaluate the results by determining the amount of the error in the sample and projecting or estimating the error in the sampling population.
  • Consider the need for further testing if the error amount is in excess of the lower limit for individually significant items for the account classification.
  • Record the projected or estimated error amount on an Error Analysis Form and consider its qualitative aspects. Determine if the causes of the sample errors indicate circumstances that could cause greater error in other items included in this or other financial statement classifications. If the error increases the risk of material misstatement elsewhere, additional substantive procedures may be necessary.

Efficiency Should Always Be an Auditor's Guide
The characteristics of a recorded population, and the relative ease of applying analytical and tests of balances procedures in an audit strategy, usually guide decisions to sample or not to sample. If, for example, an auditor can send 15 positive confirmations and cover 85 percent of the recorded population of accounts receivable, absent unusual items, or a separate class of transactions in the sampling population, a nonsampling approach would likely be most cost-efficient. On the other hand, auditing a population with 1,000 small accounts receivable having an average balance of $500, with limited variations in the balances, would probably require a sampling application to be efficient.

Statistical applications for sampling are seldom used by CPA firms except for special analytical procedures, such as regression analysis to predict expectations in account balances. Even the once-popular PPS statistical sampling method no longer has much support in practice. The reason is sample sizes determined statistically, or by statistically-based nonstatistical methods, generally are larger and require more time than using a nonsampling approach that considers risk and is based on professional judgment.

An auditor must always insist on high quality in each engagement. It is possible to produce high quality and maximize profits at the same time when auditing procedures are designed with a focus on efficiency.

More Information
These eBook resources, without CPE credit, can be obtained from my website:

  • Small Audits Made Easy and Profitable
  • Performing Auditing Tests of Balances Procedures
  • Staff Training Series for Entry-Level Accountants, New In-Charge Accountants and Engagement Leaders
  • Key Accounting Issues for Non-Profit Organizations
  • A Practical Potpourri of Time Savings on Audits
  • The Financial Reporting Framework for Small- and Medium-Sized Entities

My exclusive presentation of webcasts on and self-study courses covering various applications of auditing standards can be accessed by clicking the appropriate box on the left side of my home page, Registered users on my website receive a 20 percent discount on CPE materials presented by myself and numerous other authors on a variety of professional topics.

My assistance in CPA firm quality-control consulting, audit planning, and peer-review preparation can be obtained by sending an email using the “Contact Us” tab on my home page.

Related article:

Clarified Auditing Standards: Audit Sampling (AU-C Section 530) – Part 1


Replies (0)

Please login or register to join the discussion.

There are currently no replies, be the first to post a reply.