It is noteworthy that one area of statistical sampling in auditing that is not controversial is statistical testing and evaluation of internal controls. The reason for this unearned revenue is that the theory is uncontroversial and widely accepted and there is no evidence that individual controls testing has any problems in application. One reason MUS is the most widely used estimator, based on the Poisson distribution, is that it is a conservative estimate of the ideal or actual distribution (binomial or hypergeometric). In addition, the implementation is quite simple and can be summarized in a page or so of documentation. MUS of practice is built on these probability distributions so that sample size planning for both tests of controls and substantive tests of details is very’ similar. There has been and continues to be extensive research on statistical estimators in auditing.

Create a free account to unlock this Template

• The above is the goal of audit standards as specified by the reasonable assurance concept (ISA 200.03—05).
• Research has shown that MUS with the tainting concept has proven to be conservative in that the assurance actually obtained using MUS is greater than the nominal level specified by the confidence level.
• The auditor could then design substantive procedures to address this batch of revenue transactions instead of designating the entire control as ineffective.
• The auditor must assess each component to determine an appropriate level of audit risk and design and execute audit procedures that address the identified risks.
• But there is little current evidence of the effect on audit environments of these changes.

For instance, if an auditor is examining a ledger with thousands of entries, they might select every 50th entry after a random start, ensuring a manageable yet representative sample. Explore how different audit sampling methods influence the overall quality and reliability of audit results. At an aggregate level such processes help create new aspects or dimensions to society such as the consumer society, risk society, or even audit society (ibid., 48). The earlier social theory research reviewed in Chapter 2 seemed to be primarily concerned with the negative aspects of the auditor’s social role. Below is how the sampling risk is related to the audit tests and Retail Accounting results of the audit work. Usually, it is impossible or impractical for auditors to have time to review every record in an entire population; hence, they need to determine and rely upon a sample in performing their audit tests and making a conclusion based on the sample.

Sampling Risk Related to Audit Tests

9The auditor who prefers to think of risk levels in quantitative terms might consider, for example, a 5 percent to 10 percent risk of assessing control risk too low. In this case, there is a risk that they may accept the book value of the account when it actually contains material misstatement; and it would be detected if a bigger sample or a whole population were to be tested instead. On the other hand, they may reject internal control reliance when internal control is actually effective. This may lead to them performing more work on tests of details than necessary making the audit work not efficient. In addition, it may include inventory or revenue recognition and ongoing communication and collaboration with company management to ensure the audit is conducted effectively and efficiently. Finally, the auditor assesses the detection risk, which is low due to the use of a comprehensive audit plan, including sampling and testing of the company’s financial records and reports, as well as the experience and expertise of the audit team.

#3 – Detection Risk

The detection risk of audit evidence for an assertion failing to detect material misstatements is 5%. The audit, therefore, provides (1 – .05) assurance that the financial statements are free from material misstatement. While it might seem intuitive that larger populations require larger samples, the relationship is not always linear. In many cases, the sample size needed to achieve a certain level of confidence and precision does not increase proportionally with the population size. This is where statistical formulas and tables come into play, helping auditors determine the optimal sample size based on the specific characteristics of the population and the audit objectives. This is a very current topic in an age of crypto currencies, blockchains, artificial intelligence (Al), big data, and data analytics in auditing.

How to Reduce Sampling Risk

The study thus illustrates the key importance of controlling judgmental errors in implementing the application of the audit risk model. Further complications are introduced if, as is likely in practice, most audit procedures are performed judgmentally without using statistical models (e.g., see Hall et al. 2006). Then it can be very difficult to determine the actual level of assurance provided, and this is the feature that critics of the profession have exploited, e.g., see the critique of the vagueness of current audit standards in Power (1997) as quoted in Brydon (2019). The profession has tried to control for judgmental errors through education and experience requirements, firm- and profession-wide quality-control practices, and practice inspections.

• Legitimate assurance means that there is an appropriate argument that will convince skeptics of the appropriateness of the auditor’s conclusion.
• Stratified sampling takes the process a step further by dividing the population into distinct subgroups or strata based on specific characteristics, such as transaction type or account size.
• This practice is essential for ensuring that audits are both efficient and effective, given the impracticality of examining every transaction or account.
• The auditor then assesses the control risk, which is moderate due to the company’s implementation of effective internal controls and procedures, such as regular employee training, quality control checks, and documentation practices.
• Audit sampling is not only applied to substantive testing but is also applied to control testing by auditors.
• However, such risk can always be managed by ensuring that an appropriate sampling method is used and that the auditor has a sufficient understanding of the population that will be tested.

Hence, proper training and adequate supervision by audit risk model the senior audit staff can help reduce a significant chance of the risk to occur.

If the errors are isolated and minor, they may not significantly impact the overall conclusions. However, if the errors are pervasive or indicative of systemic issues, further investigation may be warranted. Selecting the right sample is an art as much as it is a science, requiring auditors to blend methodological rigor with practical judgment. One widely used technique is random sampling, which ensures that every item in the population has an equal chance of being selected.

Incorrect rejection happens when auditors conclude a misstatement exists when it does not, resulting in unnecessary additional auditing procedures. Audit sampling enables auditors to draw conclusions about entire data sets by examining only a portion, ensuring audits remain efficient and accurate. This is increasingly important as organizations handle larger volumes of transactions and data. This is the risk that the auditor will not detect a material misstatement, even if it exists. It is influenced by the nature, timing, and extent of audit procedures the auditor performs.