Sampling Methods: Types of Sampling and Examples

Sampling and Types of Sampling

• A sample is a small part of a population with all the qualities and characteristics of the whole population.
• According to Goodi and Hatt: A sample is a smaller representation of a large whole.

Probability Sampling

In probability sampling, the selection of a sample from the universe has an equal chance for all the members. In the case of the heterogeneous population, when samples are selected randomly but under certain conditions, it is also termed restricted sampling.

Simple Random / Unrestricted Sampling

A simple random sampling or sample is one in which each unit in the universe has an equal chance for selection. A simple random sample is drawn on a unit-by-unit basis. For example, the lottery system or assigning a number from 1 to 50 to fifty students.

Stratified Random Sampling

When the universe is highly heterogeneous, a stratified sample may be undertaken by splitting the universe into sub-populations, and a series of samples may be selected. The universe is divided into different groups and sub-groups or strata with respect to some control factors, and a sample is drawn from each stratum at random. The control factors refer to the attributes or characteristics according to which the universe is classified.

For example, to examine the effects of child abuse, the population could be stratified on the basis of gender and age, or if you are trying to find out the satisfaction level of students with their college, each department would be considered a strata and sample could be taken on the basis of equal percentage i.e., 20% from each department (strata).

Cluster Sampling

This is a process of sampling in which the sampling units are to be found in a “group” of individuals, the natural units, or mutually exclusive groups such as a household or family that comprises human beings. The samples drawn using this method are called clustered sampling. This method refers to the selection of clusters or groups instead of individual units. Clusters may and may not be equal in size. Clusters of equal size are generally the result of the planned conditions chosen by the researcher.

For example, if the researcher wants to know the sleeping habits of teenagers in a city. The city has schools where teenagers are easily found. Here, the schools are considered clusters, and the researcher chooses a random number of schools i.e., 10 schools and picks the teenage students from them.

Multi-stage

Multi-stage sampling is a method of sampling where a sample is chosen from a population in a series of phases, with progressively smaller groups being used at each stage. The process starts by dividing the population into distinct clusters, followed by the selection of a sample from these clusters. After that, smaller subgroups are discovered within the selected clusters, and another sample is generated from these subgroups.

This process persists until the last step, whereby individual components are chosen from the smallest units. This approach is especially beneficial for extensive and widely spread populations, since it streamlines data gathering and minimizes expenses in comparison to sampling each individual.

Systematic Sampling

This is a technique in which samples are drawn according to some predetermined pattern. This type of sampling is not Random, but a system is found, i.e., 0, 3, 6, 9, and 12.

For example, to evaluate students’ performance in a college, every 10^th student is selected from each class.

Non-probability Sampling

In non-probability sampling, the selection of the elements is not based on probability theory, but personal judgement plays a significant role in the selection of the sample. Elements are selected individually and directly from the population. It is also called Non-random sampling.

Purposive / Judgemental Sampling

A purposive sample is one that is selected by the researcher’s free choice and self-will.  In this method, the researcher selects samples according to the study’s needs and requirements. Such a sample is said to be biased and discriminatory in the sense of self-selection. The degree of bias is said to be large in such a sample.  In purposive sampling, the selected sample is considered the mirror of the whole in relation to the characteristic in question.

For example, the researcher wants to study child beggars from a city, and the researcher is well aware of the areas where child beggars are found. Here, the suitable sampling technique would be purposive sampling because the researcher chooses participants deliberately.

Quota Sampling

In which the information is collected from a specified number of individuals. The group of individuals is called a quota, which is specified as to sex, age, income, or other characteristics, i.e., urban and rural, old and young, upper and lower income, educated and uneducated, able and unable, and so on.

For example, a researcher wants to the study effects of social media on youth. The researcher would differentiate the youth on the basis of gender, age, and education; however, 50 males and 50 males (100), 25 respondents aged from 15-20 and 25 respondents from 20 to 25 (50), whereas 10 from each educational level i.e., matric, intermediate and bachelors (30). A total of 180 participants would be selected on the basis of quota sampling.

Convenience Sampling

It is also called accidental or incidental sampling.The research in incidental sampling picks up the cases that fall to hand, continuing the process till the size of the sampling selects the samples conveniently available from different walks of life, e.g., teachers, workers, retailers, housewives, etc. It refers to that part of the population being investigated, elected by convenience.

For example, collecting reviews from customers before leaving the restaurant is a convenient method of data collection that will provide instant data and will be helpful in improving services. In some cases, the results from this convenience sample cannot be applied to the entire population due to potential sources of bias generated by the sampling method.

Snowball Sampling

Building of a sample through informants. In this type of sampling, if a questionnaire is filled out by one group or individual, the next questionnaire is filled out by that particular individual or by group recommendation.

For example, if the researcher is conducting a study on drug addicts, the rehabilitation centers would be the primary source for data collection and then, based on their referrals, other drug addicts are approached.  

Characteristics of a Good Sample

• Greatest possible accuracy.
• Free from error due to unbiased.
• No substitution to originally selected units.
• Representative of the whole universe.
• Small or adequate in size.