Statistical sampling is an approach to sampling that involves random selection of the sample items and the use of probability theory to evaluate the sample results, including the measurement of sampling risk. Any other approach is described as being ‘non-statistical’.

Both statistical and non-statistical approaches:

Most audit testing is not performed on a rigorous statistical basis. Selection of items for testing on a genuinely random basis is often practically difficult and time consuming and the auditor may conclude that a judgemental approach to sample selection will provide the assurance required.

When using non-statistical methods, the auditor nevertheless needs to be satisfied that the selection is sufficiently representative of the population and of a sufficient size to enable him to draw reasonable conclusions regarding the population as a whole.

In both statistical and non-statistical approaches sample sizes will depend on factors such as population size, materiality and identified risk