Introduction

As per SA 530, audit sampling refers to the application of auditing procedures to less than 100% of items within a population relevant under audit such that all the sampling units have an equal chance of selection.

In simple words, sampling in auditing refers to the practice deriving a conclusion by the auditor about a population of data by evaluation of only a part or sample of the whole data. Population means a set of data.

Concept of sampling

We know, an audit involves inspection of financial information of an entity by an auditor to form an opinion on its financial statements. Now the financial information of a firm usually contains large volumes of data. For example, a firm may have entered into 50,000 purchase transactions in a year.

Now, checking each and every purchase transaction will cost both time and money. Also, nowadays, almost every enterprise have internal controls and automated accounting systems that are established to ensure accuracy and prevention of errors. Hence, a  full-fledged inspection of each and every transaction is not worth the time and effort.

Instead, a wise thing to do is to take a sample from the whole volume of transactions or accounts and apply the auditing procedures to the sample. The results derived from the sample are then projected upon the whole volume of data. Samples are often taken using statistical methods to ensure that sample is taken randomly and represents the whole population of data in a true and unbiased manner.

Consideration regarding the population before audit sampling:

1. The population is appropriate for the specific audit objective of the auditor
2. It is from a reliable source to ensure sample reliability
3. It is complete in terms of coverage of all relevant items throughout the period.

Irrespective of the method of sampling, the sample must represent the whole population closely.

Approaches to sampling

There are two approaches to sampling:

1. Statistical Approach: It is a scientific way of ensuring that the sample is chosen randomly from data and represents the data in a true and unbiased way. It employs mathematical and statistical tools like the theory of probability and also considers sampling risk characteristics.
2. Non-Statistical Approach: Under this approach, the auditor employs his personal experience to collect sample from the population. No mathematical tools are used but the personal judgement of the auditor regarding sampling. Sometimes, this approach may give satisfactory results depending upon the capability of the auditor. But in most cases, reliability is less compared to the statistical approach.