Sample Testing

Definition

The definition of sampling, as described in ISA 530 Audit Sampling is:

"The application of audit procedures to less than 100% of items within a population of audit relevance such that all sampling units have a chance of selection in order to provide the auditor with a reasonable basis on which to draw conclusions about the entire population."

The need for sampling

It will usually be impossible to test every item in an accounting population because of the costs involved. Consider a manufacturer of fasteners (i.e. nuts, bolts, nails and screws); they will have many thousands, maybe millions, of items of inventory. It would simply be impossible to test the valuation of every single one.

It is also important to remember that auditors give reasonable not absolute assurance and are therefore not certifying that the financial statements are 100% accurate.

Audit evidence is gathered on a test basis. Auditors therefore need to understand the implications and effective use of sampling.

Statistical sampling

Statistical sampling means any approach to sampling that uses:

• random selection of samples; and
• probability theory to evaluate sample results.

Non-statistical sampling

Any approach that does not meet the characteristics of statistical sampling is considered to be non-statistical sampling.

The approach taken is a matter of auditor judgement.

Designing a sample

When designing a sample the auditor has to consider:

• the purpose of the procedure;
• the combination of procedures being performed;
• the nature of evidence sought; and
• possible misstatement conditions.

The principle methods of sample selection are:

• Random selection - this can be achieved through the use of random number tables;
• Systematic selection - where a sampling interval is used (e.g. every 50th balance);
• Monetary unit selection - selecting items based upon monetary values (usually focussing on higher value items);
• Haphazard selection - auditor does not follow a structured technique but avoids bias or predictability; and
• Block selection - this involves selecting a block of contiguous (i.e. next to each other) items from the population. This technique is rarely appropriate.

When non-statistical methods (haphazard and block) are used the auditor uses judgement to select the items to be tested. Whilst this lends itself to auditor bias it does support the risk based approach, where the auditor focuses on those areas most susceptible to material misstatement. This usually leads to a focus on the higher value items within a population and is a common method in practice.

The sample size depends upon the level of sampling risk that the auditor is willing to accept.

Sampling risk

Sampling risk is a component of detection risk, the other component being non-sampling risk, and that sampling risk arises from the possibility that the auditors' conclusion, based on a sample, may be different from the conclusion that would be reached if the entire population were subjected to the same audit procedure.

Auditors are faced with sampling risk in tests of controls and in substantive procedures. Sampling risk is essentially the risk that the auditor's sample from a population will not be representative. In other words it will, by chance, include too many or too few errors to give a realistic impression of the population as a whole.

In order to reduce sampling risk the auditor needs to increase the size of the sample selected.

Rating :	Ratings & Comments  (Click the stars to rate the page)

Tags:	sampling;sampling risk;statistical sampling;non-statistical sampling