Principles of Auditing: Please read below article and can you suggest 3-4 points

Question

Principles of Auditing: Please read below article and can you suggest 3-4 points from it that should be considered to avoid bias in sample selection?

Auditors: Use a Clear Sample: By Richard E. Wortmann, CPA

Before any discussion on audit sampling, you may want to open up AU Section 350, Audit Sampling, to review the definition. Audit sampling is "the application of an audit procedure to less than 100 percent of the items within an account balance or class of transaction for the purpose of evaluating some characteristic of the balance or class." Many procedures that auditors perform use sampling methodologies. Unfortunately, there are instances where auditors think they are sampling when, in fact, they are not. They do not close the loop on sampling procedures and neglect to draw the appropriate conclusions. AICPA issued a revised Audit Sampling Guide in 2008, and AICPA revised AU Section 350 with Statement of Auditing Standard Number 111.   

A common error in the application of audit sampling is selecting a sample from a population that is not the complete population the auditor is trying to conclude. For example, a sample of payroll transactions from one pay period or cash disbursements from one month. These examples would only permit an auditor to conclude on that one pay period or that one month of cash disbursements. The above methodology is considered "block sampling," and though it is common, it is not appropriate for audit purposes.   

Auditors are cautioned that certain tests are not sampling procedures; therefore, the results of the tests cannot be extrapolated. Examples include inquiries, analytical procedures, procedures applied to all items in a population, and classes or transactions not subject to selection or possibility of selection.  

Sampling techniques require that every transaction in a population being sampled must have an equal opportunity as any other transaction to be selected. Therefore, to conclude on payroll for the year, every transaction from the year must have an opportunity to be selected. Audit sampling does not prescribe how the sample should be selected - random, haphazard, systematic - but standards do caution that the selection method should be without bias.   

When performing attribute testing used for control testing, auditors must not weigh their selections toward large dollar items. The results of attribute sampling are a rate of errors that permit the auditor to rely, or not rely, on the controls tested. Dollar amounts become important when performing variables sampling and when the conclusion is on the accuracy of the dollar amount presented. It is not unusual to find the auditor stratifying the sample between large items, which are tested 100 percent, and the rest of the population, which is tested by sampling. When this is the case, the auditor must not use the results of the items reviewed 100 percent to conclude on the rest of the transactions. The large items are not being tested using sampling methodologies. Further, the results of the smaller items that were tested by sampling methodologies must not be projected to the items tested 100 percent since the large items were not in the population.  

Another common error in audit sampling is inappropriate sample sizes. Sample sizes are, in simple terms, a result of a combination of sampling risk and precision. A revision to SAS 39 contained in SAS 111 clarified that sample sizes used in nonstatistical sampling must approximate the size that would have resulted if statistical sampling was used. However, the standard also clarified that the auditor does not need to calculate a statistical sample to make sure the nonstatistical sample size is valid. What the auditor does need to do is use valid sampling models to ensure compliance with the standard. If the auditor is using third-party practice aids, the underlying guidance should be reviewed to make sure the firm is comfortable with the risk tolerances adopted by the authors. If auditors develop their own sampling tools, they need to document the underlying criteria. There are many published tables to assist an auditor with developing sample size guides based on the assessed risks.   

For auditors who use data extraction software, many of these programs have sampling formulas incorporated into the systems. The auditor enters the sampling criteria, and the software calculates the statistical sample size.  

A last word of caution when selecting samples: many auditors prefer to select one sample for both attribute and variables testing, commonly referred to as a dual purpose test, since it is often more efficient to select one sample than multiple samples. To determine the appropriate sample size, the auditor must select the greater of the sample size that would be required if independently tested.

Explanation / Answer

If entire segments of the population are excluded from a sample, then there are no adjustments that can produce estimates that are representative of the entire population. But if some groups are underrepresented and the degree of underrepresentation can be quantified, then sample weights can correct the bias. Define inclusion and exclusion criteria by PICOTS clearly and in a protocol. Reduce ambiguity as much as possible. Dual review can help reduce random error in applying inclusion and exclusion criteria. Define interventions with specificity such that they are applicable to the intended user of the review.

Related Questions

Navigate

• Browse (All)
• Browse P
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.