Discovery sampling is worth discussing, as it is frequently used to detect a fraudulent transaction.

A sample size is selected for discovery sampling. If the sample is error free, then the entire data set or population is accepted as error free. However, even if only a single error is discovered, then the entire population is rejected. The auditor is not as much interested in determining how many errors there are in the population as in the fact that there was an error. The auditor's concern would be that there is a possibility that the internal control system could be compromised. The one error may be sufficient to require an examination of the entire population or to formulate a new plan for further action. If one instance of fraud is found, the auditor is 100 percent confident that fraud exists.

On the practical side, deviations must be divided into categories as to whether the deviations are:

- Clearly errors

- Clearly fraudulent

- Highly suspicious for fraud

While the goal of discovery sampling is for not even a single critical deviation, an expected deviation rate of 1 percent is typically accepted.

FIGURE 4.25 Mean Estimation Report Example of One of the Six Types Available

Discovery sampling's stringent requirement regarding errors makes it a good choice for fraud-detection audits.

Choice of Sampling Methods

An auditor first must decide whether to utilize statistical or nonstatistical sampling and then which method within the two choices.

There is much disagreement as to whether statistical or nonstatistical samples are more defensible, especially in court. Auditors tend to favor statistical sampling as sampling is independently and objectively determined. Measurements are mathematically calculated providing results that are quantitative. Auditors and accountants are comfortable with defined procedures and standards that produce specific numbers.

However, lawyers call upon expert witnesses to provide professional judgment and opinions. As such, it may be that nonstatistical sampling may produce better results as it is based on the judgment and experience of the auditor. The expert witness would explain to the court how, in their professional opinion, the sample is representative of the population. This may be more effective than having a statistician explain that for a 90 percent confidence level, the auditor's conclusion has a 10 percent risk of being wrong.

If statistical sampling is selected, then the type of statistical sample must be decided upon. The nature of the audit test should dictate the choice. If the primary purpose of the sample is to test a control, then attribute sampling is generally the best choice. CVS, while normally requiring larger sample sizes, is good for verifying account balances or a specific class of transactions. For auditing items where errors tend to be low, such as inventory or accounts receivable, PPS or MUS are better choices, as they require smaller sample sizes. The auditor must be aware that PPS has a higher probability of showing false positives due to its sensitivity of errors in projecting misstatements. For fraud detection, clearly the larger concern over any errors is the best selection.

Found a mistake? Please highlight the word and press Shift + Enter