Audit Sampling


Audit sampling is defined as the application of an audit procedure to less than 100% within a population of the audit. Population refers to any group of records or documents in the audit relevance that belongs in a specific category.

Audit sampling is usually performed in a way that all sampling units in the population have a chance of selection in order to provide us with a reasonable basis on which to make conclusions about the entire population.

It is impossible to review every record of the client. Hence, auditors need to determine how many records they should look at and what kind of records are more important to examine. This is where the audit sampling comes in to play an important role as auditors need to determine what is their sample of the records and documents to review.

Purposes of Audit Sampling

Auditors usually perform the audit sampling on both the test of controls and substantive audit procedures. In the test of controls, auditors use the sampling method to evaluate whether the client’s internal controls work effectively. In substantive audit procedures, they use the sampling method to estimate the amount of misstatement in an account balance.

Also, auditors usually design audit sampling that provides them a reasonable basis to draw conclusions about the population from which the sample is selected. Hence, it saves them a lot of time and effort comparing to auditing the entire population of an account balance.

Usually, in an applicable auditing standard, e.g. in ISA 530, auditors need to follow the audit sampling guideline, so performing audit sampling can be considered as a basis of compliance with the required auditing standard.

Moreover, the audit sampling method is usually used when auditors want to make an investigation on a specific matter that comes up when performing the audit work.

In summary, purposes of audit sampling include:

  • To gather audit evidence to make a conclusion on the client’s controls and account balances
  • To reduce the work done but still provide a reasonable basis to draw conclusions
  • To comply with an applicable auditing standard
  • To be used as a tool to investigate a specific matter that could lead to misstatement which is due to error or fraud.

2 Types of Audit Sampling

Statistical audit sampling and non-statistical audit sampling are the two common approaches of audit sampling. Both statistical and non-statistical audit sampling can provide auditors a reasonable basis to draw conclusions on the population in order to form an audit opinion. Either approach of audit sampling can provide auditors sufficient audit evidence when applied properly.

Statistical Audit Sampling

Statistical audit sampling is the sampling method that involves the application of probability theory and statistical inference.

Auditors usually statistical sampling along with their professional judgment to assist them in deciding an appropriate sample size and evaluating the sample result.

Auditors usually apply statistical sampling by using tables or formulas to compute the sample size they need to select. It allows them to control and measure the risk of making an incorrect inference about the population from which the sample is taken.

In statistical sampling, each item in the population has a calculable chance of being selected as a sample to be examined.

A commonly held misconception about statistical sampling is that it removes the need for auditors’ professional judgment.

However, this is not true because before the sample size can be determined, auditors need to use their professional judgment to consider many factors such as materiality, the expected error rate or amount, the risk of over-reliance or the risk of incorrect acceptance, audit risk, inherent risk, control risk, standard deviation, and population size.

For example, auditors have 500 items in the population and need to select a sample to examine. In statistical sampling, after using their professional judgment to consider the above factors, auditors determine to have 10% as the sample sizes which are 50 items.

In this case, they may select 50 items randomly from the population or they may decide to select items from a fixed interval from the first item to the second item and onward. Hence, any item among the 500 has a chance to be selected.

Non-Statistical Audit Sampling

Non-statistical audit sampling is the sampling method that auditors do not rely on the application of probability theory or statistical inference. Also, they do not use tables, formulas, or statistical percentages to compute the sample size.

Instead, auditors rely solely on their professional judgment in determining the sample size and selecting the items to sample. The number of records or items they select for examination will be based on their professional judgment and experiences as well as the policy in their firm.

It is useful to note that if auditors use non-statistical sampling to determine the sample size and select the sample to test, the sampling risk cannot be measured or controlled. Also, each item in the population does not have an equal chance to be selected into the sample at all.

For example, the firm policy may state that if the control risk is high, auditors need to select from 100 to 150 records. If control risk is moderate, they may select from 50 to 100 records and if the control risk is low, they can select from 20 to 50 records to be examined.

However, this is just an example; in many cases, the policy in the firm is usually just a guideline and the different number of records may be selected based on auditors’ professional judgment. Auditors’ decisions are based on many factors such as inherent risks, control risks, audit risks, and population size, etc.

5 Types of Sampling Methods

The five types of sampling methods include random selection, systematic selection, monetary unit sampling, haphazard sampling, and block selection.

Sampling Methods

Random selection

Auditors use random selection by using random number tables or random number generators. With this method of sampling, auditors can ensure that all records or items in the population have an equal chance of being selected.

Systematic selection

Auditors use a systematic selection method by setting up a fixed interval for items to be selected. The first item can be randomly selected and then from the second item onward will be selected based on the set fixed interval.

For example, if they set the interval to be 10 and the first item selected is 8th item, then 18th, 28th, 38th, 48th item and onward will be selected as the sample items.

Monetary unit sampling

This method of sampling is a type of value-weighted selection which is based on monetary amounts. The sample size, the sample selection and the evaluation of the result will be based on the monetary amounts.

Haphazard sampling

Haphazard sampling is the method of sampling that does not follow any structured technique. In this method, auditors choose a sample of records in a population without any conscious bias.

Though the name is “Haphazard”, auditors should make sure to select the records that represent the entire population.

Block selection

As the name suggested, the items selected for a sample in this method are based on the block.For example, auditors may decide that all items in the month of January or the invoices with the number from 00100 to 00199 will be selected as the sample items.

In this method, auditors select a block of contiguous items from within the population as a sample. Hence, it is rarely an appropriate sample selection technique for auditors when they intend to draw valid inferences about the entire population based on the sample.