Audit Sampling Concepts
• The only way we test balances (especially with the income statement)
The Nature of Sampling
• More cost effective and efficient than taking a census
• Auditing uses sampling to deal with three key aspects:
o Planning the sample
o Selecting the sample and performing tests
o Evaluating the results
• A representative sample is one where the characteristics of the sample are approximately the
same as those of the population
• A non-representative sample is one where the characteristics are not approximately the same
as those of the population.
• Non-sampling risk arises when the auditor uses inappropriate audit procedures, fails to detect a
misstatement when applying an audit procedure, or misinterprets the audit results
o It is reduced by adequate training, proper planning and effective supervision
• Sampling risk results from testing less than the entire population. It is the risk that conclusions
based on the sample results will not be true of the population as a whole.
o Can be reduced by increasing the sampling size
• Statistical and non-statistical sampling refer to the type of sampling method
• Probabilistic and non-probabilistic sampling refers to the way the sample is chosen and tested
• Statistical sampling uses the laws of probability for selecting and evaluating a sample from a
population for the purposes of reaching a conclusion about the population that has been
selected at random
o The found misstatement is then projected on the entire population
o The auditor is able to quantify the sampling risk
• Statistical sampling also gives rise to two types of errors:
o Type 1 Error – the risk as assessing the control risk as too high/risk of incorrect
rejection. The auditor may perform more work than necessary
o Type 2 Error – the risk of assessing control risk too low / Risk of incorrect acceptance –
the auditor may fail to detect a material misstatement.
• Statistical sampling has a number of advantages in that it provides for:
o Quantitative evaluation of the sample results.
o A more defensible expression of the test results.
o More objective recommendations for management.
• When statistical sampling is used, the sample must be probabilistic, and appropriate statistical
evaluation methods must be used with the sample results to determine sampling risk.
• In non-statistical sampling the auditor uses subjective judgement in determining the sample
size and selection process to the audit items of greatest value and highest risk
o Conclusions are reached about the population based on a judgemental basis
o A non-statistical sample can be designed so that it is equally effective and efficient as
statistical sampling — while being less costly.
• Non-statistical sampling does not allow the auditor to draw objectively valid statistical
inferences from the sample results, or quantitatively measure and express sampling risk. • When using probabilistic sample selection, the auditor randomly selects items such that each
population item has a known probability of being included in the sample.
• When using non-probabilistic sample selection, the auditor selects sample items using
professional judgment rather than probabilistic methods.
Process of Sampling
• Whether the sample is statistical or non-statistical, the same process applies
• This will be discussed later on in the chapter (14 steps)
o Plan the sample to make sure that the audit tests are performed in a manner that
provides the desired sampling risk and minimizes the likelihood of non-sampling error.
o Perform the tests by examining documents and performing other audit tests.
o Evaluate the results by drawing conclusions based on the audit tests.
o Select the sample by deciding how to select sample items from the population.
Non-Probabilistic and Non-Stastical Samples
• Directed sample selection — each item in the sample is selected on the basis of some
judgmental criteria established by the auditor. Refer to Table 11-2.
• Block sample selection — the auditor selects the first item in a block, and the remainder of the
block is chosen in sequence.
o Test every sample within a block
• Haphazard sample selection — where the items for the sample are chosen without regard to
their size, source, or other distinguishing characteristics.
Probabilistic and Statistical Samples
• The auditor may choose to stratify the population before selecting a sample.
• Simple random sample selection — one in which every possible combination of population
items has an equal chance of constituting the sample.
• Systematic sample selection — where the auditor calculates an interval and then methodically
selects the items for the sample based on the size of the interval.
• Probability proportionate-to-size sample selection — a variation of systematic sample selection
by unit of interest where the individual dollar is considered the unit of interest.
o The interval is determined based upon a statistical formula, and the transactions
associated with that dollar interval are selected.
• Once the decision has been made to conduct statistical sampling, the auditor may choose from
three broad categories of statistical sampling:
o Probability-proportionate-to-size (PPS)
• Attribute sampling is typically used for tests of controls — a sample is drawn to estimate the
proportion of an attribute or characteristic in a population