What is attribute sampling?
Attribute sampling is a statistical process used in auditing procedures aimed at analyzing the characteristics of a particular population. This practice is often used to test whether a company’s internal controls are being properly followed. Without the ability to rely on controls, performing meaningful audits can be very difficult and very costly.
- Attribute sampling is used in auditing procedures to help analyze the characteristics of a particular population.
- This statistical process helps determine if internal controls are being followed.
- This process takes a sample of a particular activity, such as paying a certain amount of invoice and analyzing the process.
How attribute sampling works
An auditor wants to test the validity of the company’s rule that purchases over $ 10 must be approved on the purchase order. Auditors take samples because it is often infeasible to look at invoices from all vendors over $ 10. The size of the sample should be large enough to provide an accurate image of the entire population of purchase orders over $ 10, but its accuracy is always a matter of degree and should be tested. .. Examining the sample, auditors may find that 5% of vendor invoices over $ 10 are not approved by the purchase order.
On the other hand, 5% may be considered acceptable. The auditor was unable to take the sample and examine the entire vendor invoice population, resulting in a phenomenon called a “sampling error” each time the sample was taken, requiring additional analysis. ..
Sampling errors occur when the values in the sample do not match the values in the entire population from which the sample was taken. So if additional analysis shows that the margin of error is 2.5%, then the confidence interval is 5% plus or minus 2.5%, and the acceptable 3% rate is within that range, so 5%. Nonconformity rates are acceptable.
This 5% nonconformity rate may or may not be acceptable, depending on the rate the auditor deems acceptable. If the auditor believes that a rate of 3% is acceptable, then 5% appears to be too high as a result, indicating that the company’s internal controls were ineffective. This scenario requires additional investigation by the auditor. This data may also suggest that additional management is needed in the future.
Consider an election poll. According to sample data, 49% of those surveyed will vote for Candidate A, and 51% of those surveyed will vote for Candidate B. In this case, there is a sampling error of 2.5%. Is greater than the difference of 2% between the numbers, which makes the voting result in question unreliable.
Attribute sampling only makes sense when used to audit well-designed and efficient internal controls.
Types of questions asked in attribute sampling
Many items can be investigated using attribute sampling. The partial list includes:
- Is the item always shipped before billing?
- Will the product be purchased only from approved vendors?
- Will credits be sold to consumers with approved credits?
- Are all checks signed by an authorized person?