CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. 78.1% of retail investor accounts lose money when trading CFDs with this provider. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.
US English

What is sampling?

By Fitri Wulandari

Reviewed by Vanessa Kintu

Fact checked by Richard Reed


Sampling is the process or technique of selecting a subset of a larger group to assess the group’s characteristics. Getting information from a huge data set might take a long time, so sampling data is a faster way to get equivalent results. 

How sampling works?

The meaning of sampling is randomly selecting data from a population to produce a result that is similar to or identical to the result produced by the entire population. However, there are differences between the sample result and the population result. This is known as sampling error, and it can be avoided by not generating a sample but instead, considering the entire population set.

Where is sampling used?

Sampling can be used for research in accounting. For instance, a certified public accountant (CP) can use sampling to determine the accuracy of account balances in the financial statements during an audit. 

Types of sampling

Based on the sampling definition by the Corporate Finance Institute, there are two types of sampling in statistics: probability sampling and non-probability sampling. The sampling methods used are determined by the type of analysis being conducted.

Probability sampling

In the probability sampling method, every member of the group has an equal chance of being chosen for the sample.There are four types of probability sampling:

  1. Simple random method

It is the least prone to bias because there is no human judgement involved in the selecting of the sample. For instance, choosing the names of 20 students out of a school with 100 students. Each student has the opportunity to be selected for the sample.

2. Systematic method

Assigns numbers to individuals or items. They are chosen based on the numbers at a defined, regular interval. By dividing the population size by the sample size, the sampling interval is obtained. 

A systematic sampling example could be an auditor wanting to check a company policy that staff employed for less than one year must have annual leave approved by two team leaders. The population would consist of new employees, let’s say 100. With 20% of sample size of the population or 20, the sampling interval would be five (100 checks/20 sample checks). The auditor will check every fifth employee.

3. Stratified method

Divides individuals into sub-groups based on certain criteria, such as profession, age or gender to ensure each of the sub-groups is represented in the sample. 

For example, a company wants to check gender ratio. There are 70 female employees and 50 male employees, who are divided into two sub-groups based on gender. Then, the company can pick 35 female and 25 male employees. 

4. Cluster method 

Also divides a bigger population into subgroups, but the divisions are chosen at random. A corporation that owns many plants across the country with a similar number of workers in each plant is an example of this strategy. At random, the company just needs to visit three to four factories. Instead of visiting each office, they are chosen at random to achieve the desired results.

Non-probability sampling

With this type of sampling, not all individuals or items are selected. Types of non-probability sampling include:

1. Convenience method

Includes people who are readily available to the researcher. The method is simple, quick, and inexpensive. For example, a professor who wishes to evaluate his or her performance can ask his or her students to complete a survey shortly after the lecture. This is the most practicable method for gathering data from all students.

2. Voluntary sampling method

Individuals volunteer to participate in the survey rather than the researcher picking the participants, as the name implies. For example, an electronic store’s customer care department may ask customers to complete a survey after they make a purchase. However, this can introduce bias, as certain types of customer (eg very happy or very unhappy) are more likely to volunteer to take part in the survey.

3. Purposive method

The researcher selects respondents who are suitable for the topic of research. For example, a railway company wanting to know about access for disabled passengers will choose passengers with disabilities to take part in the poll.

4. Snowball method

When it is difficult to find respondents, a researcher can ask one respondent to help to connect with other individuals who can participate in the survey. For example, if the researcher wants to look into issues faced by migrant workers in a country, they find one individual who can connect them with other migrant workers. 

Latest video

Latest Articles

View all articles

Still looking for a broker you can trust?

Join the 610,000+ traders worldwide that chose to trade with

1. Create & verify your account 2. Make your first deposit 3. You’re all set. Start trading