Chapter 7: Introduction to audit evidence
© EWP Go to www.emilewoolfpublishing.com for Q/As, Notes & Study Guides 143
Audit sampling: ISA 530
The nature of sampling
Sampling risk and statistical sampling
Sample design
Performing audit procedures on the sample
Projecting misstatements and evaluating the results of audit sampling
3 Audit sampling: ISA 530
3.1 The nature of sampling
As discussed in the previous chapter, auditors do not normally check 100% of
transactions and balances that go into the production of financial statements.
However, sampling is not always appropriate for auditing. For example, if a
population consists of a small number of large items, it may well be appropriate to
apply audit tests to the entire population. Also, sampling is only appropriate where
the population is homogeneous (i.e. where all the items in the population share
common characteristics). If this condition is not satisfied, it may be necessary for the
auditor to test the entire population.
For example, suppose that a company owns just three very large machines, and each
machine serves a different purpose. It would be appropriate to carry out audit tests
on all three machines instead of testing just one or two machines as a ‘sample’ of the
entire population of three machines.
Again, as discussed in the previous chapter, sampling in auditing involves applying
audit testing procedures to less than the entire population of items subject to audit.
So, if a company has issued 1,000 sales invoices in a period and the auditor tests 999
of them, this is a sampling exercise – although the size of the sample would be very
large, given the size of the total population!
The auditor is not interested in the results of the sample itself. The sample is used as
a basis for reaching a conclusion on the entire population. For example, if an auditor
tests 500 sales invoices out of a population of 1,000 invoices, the purpose of the audit
test is to make a conclusion about all 1,000 invoices, not just the 500 that are tested.
It is therefore important that the sample chosen should be representative of the
population, and should reflect the characteristics of the population as a whole. The
larger the sample, the more likely it will be to be representative and to reflect the
characteristics of the entire population. However, large samples are more time-
consuming for the auditor to select and test. For reasons of time and cost, the
auditor will want to test as small a sample as possible, consistent with the aim of
achieving the required level of detection risk and limiting audit risk to a particular
level.