Sampling Methods: Techniques, Types, Examples

Sampling is a statistical procedure of drawing a small number of elements from a population and drawing conclusions regarding the population.

What is Sampling?

Sampling is the process of selecting a subset of people or social phenomena to be studied from the larger universe.

The main objective of sampling is to draw inferences about the larger group based on information obtained from the small group.

The main way to achieve this is to select a representative sample. A sound representative sample should reflect all variables that exist in the population.

The term ‘population’ refers to all those who could be included in the survey. A variable is any characteristic on which people or groups differ.

A variable is a set of mutually exclusive attributes of a sample unit: sex, age, employment status, etc. The elements of a given population may be described in terms of their attributes on a given variable.

The variable is closely associated with the term sampling frame. The sampling frame lists all units in the population from which the sample will be selected.

The sampling method is less expensive and less time-consuming than the census technique. It is convenient to administer a sample method as the small sample units can be easily manageable.

The sampling method is also useful for the intensive and elaborate study of selected units.

The main assumption behind the sampling technique is that though socio-legal phenomena are complex, there appears to be a dominant unity in diversity, and it is possible to draw a representative sample.

However, the choice of the unit should be clear, unambiguous, and definite. Moreover, the sample unit must be adequate in size to be reliable.

However, to be reliable, the choice of sample units should be made with due care, and the subject matter under the survey should be homogenous.

The main advantage of the sampling method is that it can facilitate the estimate of the characteristics of the population in a much shorter time than would be possible otherwise.

It is also less expensive as only fewer people need to be interviewed.

However, the sampling method also has some disadvantages, such as the possibility of biases in selecting units, leading to false conclusions.

Biasness occurs when the decisions of the researcher about whom to sample are influenced too much by personal judgments, prospective respondents’ availability, or his implicit criteria for inclusion.

A biased sample does not represent the population from which the sample was selected.

The use of sampling methods also requires the knowledge of sampling and the selection of appropriate samples.

Moreover, if the units under sampling are liable to change, it isn’t easy to maintain homogeneity.

What is Population in Research Sampling?

A population (also called a universe) is the total collection of all the population elements, each of which is a potential case.

All students in a college, for example, constitute a population of interest, and each student in the college questioned about his/her age, height, weight, or opinion on any issue is a population element.

What is Census?

A census is an investigation or a count of all the population elements. Any part of the population is a sample. If a sample is selected according to the rules of probability, it is a probability sample or random sample.

If a sample is random, it is possible to calculate how representative the sample is of the wider population from which it was drawn. The counterpart of the probability sample is the so-called non-probability sample.

What is Non-probability sampling?

Non-probability sampling is a non-random and subjective method of sampling where the units’ selection depends on the sampler’s personal judgment.

What is a Survey?

A survey is a general term that refers to the collection of data using interviews, questionnaires, or observations.

What is Sample Survey?

A sample survey is a study involving a subset (or sample) of individuals selected from a larger population by accepted statistical methods.

What is Target Population?

A distinction is sometimes made between the target population and the sampled population. In research, the target population is the entire set of units for which the survey data is used to draw conclusions and make inferences.

It can also be defined as the eligible population included in research work. It is also called the survey population.

Ideally, the two will be the same, but for practical reasons, there will usually be differences between them.

One reason for the difference is that some of the population in a survey are not covered, so no information is obtained for them.

Depending on whether the sampling units are finite or infinite, a population may be finite or infinite.

What is Finite Population?

A finite population contains a countable number of sampling units, for example, all registered voters in a particular city in a given year or all customers who visited the city store in May 2006.

What is Infinite Population?

An infinite population consists of an endless number of sampling units, such as the number of coin tosses until a head appears. Sampling designed to produce information about particular characteristics of a finite population is usually called survey sampling.

What is Sampling Unit?

A sampling unit or simply a unit is a well-defined, distinct, and identifiable element or group of elements on which observation is made.

In some studies, an individual in a household may be a sampling unit, while in another study, the household may be a sampling unit. A sampling frame is a list of units or groups of units in the population to be sampled.

What is Sample Size?

Sample size refers to the number of units contained in a sample, while population size is the number of units that constitute the population.

The population characteristics about which the inferences are made are called parameters.

For a given sample design, an estimator is a method or formula for estimating the value of the parameter.

An estimate is the numerical value of the estimator obtained from the sample. Bias is a term that refers to how far the average value of the estimator lies from the parameter.

How Sampling and Sample Design Related

Sample design refers to the plans and methods to be followed in selecting a sample from the target population and the estimation technique vis-a-vis the formula for computing the sample statistics.

These statistics are the estimates used to infer the population parameters.

Implicit in the concept, the sampling design also includes issues such as the choice of the sampling frame, determination of the sample size, estimation of reliability of the estimates, stratification procedure, sample allocation method, clustering of the sample, etc.

How Sampling and Survey Design Related

Survey design is preparing a complete plan of operations to be followed in conducting a survey and disseminating its intended results.

We will elaborate on designing a survey in a separate article.

We, however, emphasize that the survey objectives covered under survey design determine the sample design. In practice, the sample design must be developed as an integral part of the overall survey design.

Survey design and sample design are thus two interrelated concepts, and one is complementary to the other.

It is almost always desired that a sample design be evaluated for its perfection, and a perfect sample design is expected to meet certain criteria, which include, among others, the criteria of accuracy, reliability, validity, and efficiency.

Importance of Sampling

A sample is taken almost always to provide statistical data on an extensive range of subjects for both research and administrative purposes.

The following examples are designed to illustrate the importance of sampling in real life:

1. In an opinion poll, a relatively small number of persons are interviewed, and their opinions on current issues are solicited in order to discover the attitude of the community as a whole.
2. Marketing and advertising agencies conduct countless inquiries to determine customers’ expectations, attitudes, buying habits, or shopping patterns. This information is useful to the manufacturers of goods for sales promotion. Since it is impossible to procure this information from countless customers, it is achieved through interviewing a part of the customers.
3. Large lots of manufactured products are accepted or rejected by purchasing departments in business or government following inspection of a relatively small number of items drawn from these lots.
4. At border stations, customs officers enforce the laws by checking the effects of only a small number of travelers crossing the border.
5. A department store wishes to examine whether it is losing or gaining customers by drawing a sample from its list of credit card holders by selecting every tenth name.
6. Auditors often judge the extent to which the proper accounting procedures have been followed by examining a small number of transactions selected from a large number of such transactions taking place within a specified period of time.
7. Ministry of Health and Family Welfare might be interested to know the status of knowledge among the adult population in Dhaka city on the danger of environmental pollution by interviewing a few selected adults of the city.

Countless measurements of the economy, health, labor force, contraceptive use, immunization, unemployment, income, export, import, industrial products, and the like rely on samples rather than on complete enumeration.

Numerous surveys are being conducted to develop, test, and refine hypotheses in sociology, psychology, demography, political science, anthropology, geography, economics, education, and public health.

Both local and central governments make considerable use of survey data to be aware of the various population characteristics for planning and development purposes.

In every case, a sample is selected because it is impossible, inconvenient, slow, or uneconomical to monitor the entire population.

It is now widely agreed that a sample survey is a popular and scientific data collection method.

Types of Sampling

Probability Sampling

It refers to a sample that has been selected using random selection so that each unit in the population has a known chance of being selected.

In other words, individual units are chosen from the whole group, not deliberately but by some mechanical processes.

Thus, probability sampling is also known as ‘random sampling.’

Probability sampling is instrumental when researchers want precise, statistical descriptions of large populations- for example, the percentage of unemployed populations or plans to vote for candidate X, etc.

Thus, probability sampling is used in large-scale surveys. Probability sampling has the advantage of eliminating human biases in sampling. The sample error in this method can be kept to a minimum.

Probability sampling enhances the representativeness of sampling and provides for generalization from a sample to the population.

There are three types of probability sampling methods are (1) Simple Random Sampling, (2) Stratified Random Sampling and (3) Non-Probability Sampling.

1.1) Simple Random Sampling

This is the basic form of a probability sample. In this random sample, each population unit has an equal probability of inclusion in the sample.

The key steps of devising a simple random sample include defining the population, deciding on sample size, and selecting the mechanical process.

Generally, in this type of sampling, the units composing a population are assigned numbers.

Then a set of random numbers is generated and the units having those numbers are included in the sample. Simple random sampling is free from bias and is generally more representative.

1.2) Stratified Random Sampling

Random sampling will likely, by chance, include a higher proportion of one group of people than there should be for it to be truly representative.

To avoid this problem, stratified random sampling is employed. Stratified random sampling is employed when the population from which a sample is drawn does not constitute a homogenous group.

Thus, stratification means grouping the units composing a population into homogeneous before sampling. In this type of sampling, the population is stratified by criteria, and then the selection is made through simple random sampling from the resulting strata.

In other words, under this method, the population is divided into several subpopulations that are individually more homogenous than the total population.

Then the selection is made from each stratum to constitute a representative sample.

The stratified random sampling ensures that the resulting sample will be distributed similarly to the population in terms of the stratifying criterion.

Stratified sampling can ensure greater representativeness of the sample if the stratification process is based on objective criteria.

1.3) Systematic Sampling

In systematic sampling, the population is listed so that its order can uniquely identify each population element.

The list of elements in the population is usually ordered randomly concerning the trait to be measured. In this sense, it is also equivalent to simple random sampling.

However, here sample is selected at every sampling interval.

Typically, simple random sampling requires a list of elements. When such a list is available, researchers usually employ systematic sampling.

For instance, if the list contains 10,000 elements and the researcher wanted a sample of 1,000, he should select every tenth element for his sample.

Non-Probability Sampling

Non-probability sampling means a sample that has not been selected using a random method. In this method, units for the sample are selected deliberately by the researcher.

Thus, in non-probability sampling, the researcher purposely chooses the particular population units with certain characteristics for constituting a sample because such units will represent the entire population.

There are two main types of non-probability sampling are (1) Judgement of Purposive Sampling and (2) Quota Sampling.

2.1) Judgement of Purposive Sampling

In the judgment of purposive sampling, the researcher selects the units to form his sample on his own judgment.

The essence of this method is that the researcher, presumably having sufficient knowledge about the population and its elements, uses his experience to select a sample that will be the most useful or representative.

This technique is useful in cases where the whole data is homogeneous, and the researcher has full knowledge of the various aspects of the problem.

2.2) Quota Sampling

This combines judgment and probability procedures. Here, the population is classified into several categories based on judgment, assumption, or previous knowledge.

First, people are selected globally: gender, age, class, locality, etc.

For instance, in conducting research, the researcher may need to know what proportions of the population are male and what proportion are female, what proportions of each gender fall into various age categories, educational levels, ethnic groups, etc.

Quota sampling aims to produce a sample that reflects a population in terms of the relative proportions of people in different categories.

Quota sampling is much quicker and cheaper than proper probability sampling.

Sampling With and Without Replacement

A sample may be drawn with replacement (SWR) or without replacement (SWOR).

Suppose the sample is taken with replacement from a population, finite or infinite.

In that case, the unit drawn is returned to the population, and the number of units available for future drawing is not affected.

Consequently, the probability of drawing any remaining unit in successive selections will remain unaltered.

In sampling without replacement, the unit drawn is not returned to the population in subsequent drawings. Unlike sampling with replacement, the probability of drawing any remaining unit in successive selections will be increased.

Sampling with replacement is sometimes referred to as unrestricted sampling. In general, sampling with replacement is less precise than sampling without replacement.

Intuitively, sampling with replacement seems rather wasteful.

In practice, almost all sampling is done without replacement since there is little justification for studying the characteristics of the units, which have already been included in the previous selection.

Sampling with replacement is of interest primarily for theoretical interest since the formula for the variance and estimated variance of the estimators are often simpler when the sampling is made with replacement than when it is made without replacement.

Example#1

Suppose we have 4 members in a family to whom we assign serial numbers 1, 2, 3, 4.

We need to select two of them for an interview. In this particular instance, we say that we have a population of size 4 (i.e., AM) from which a sample of size 2 (i.e., m=2) is to be selected.

To select these two members without replacement, there will be altogether 6 possible samples of each of 2 members. The accompanying table displays all possible samples of size 2:

Table: Samples of Size 2 without Replacement

Sample #	Samples consisting of Serial numbers
1	(1.2)
2	(1,3)
3	(1,4)
4	(2,3)
5	(2,4)
6	(3,4)

In practice, we deal with only one sample, which might be any of the above cases.

If sample number 3 is selected, we will seek information from members 1 and 4 to meet our survey objectives.

Example# 2

Refer to the Example above. If sampling is done with replacement, there will be 16 possible samples, each of size 2. Table 5.2 shows these samples.

Table: Samples of Size 2 with Replacement

Sample #	Serial numbers	Sample #	Serial numbers
1	(1-1)	9	(3,1)
2	(1,2)	10	(3,2)
3	(1,3)	11	(3,3)
4	(1,4)	12	(3,4)
5	(2, 1)	13	(4,1)
6	(2, 2)	14	(4, 2)
7	(2, 3)	15	(4,3)
8	(2, 4)	16	(4, 4)

Note that sample 3 (for example) and sample 9 are, in fact, identical. If we select sample 3, we do not obtain any extra information by selecting sample 9.