## Introduction

Sampling is a crucial aspect of any research process hence it has to be planned as per the needs of the study. In social work research too it is a significant part which gives authentication to your research. The analysis and the accuracy of the research is determined by the appropriate sample size and technique. Let us now try to understand the different terminology used in sampling process.

The term **sample** refers to a small part from the whole which is a representative of the qualities of the larger whole.

**Sampling** is a methodology in which the exact representation of the whole is determined through a small part of the whole.

It is an important aspect not only in research but in our daily lives too. You all must have noticed the process of taking a sample while cooking. It gives an idea whether the food is ready or it needs modifications.

Now let us try to understand it in terms of social work research.

## Sampling

When carrying out research it is not possible to study the whole population especially if it is of large number. Therefore, a methodological approach is applied to determine a small portion which would represent the whole population; and there are different ways and techniques of sampling in research depending upon the type of population to be studied

In socialwork research also it is a procedure by which a representative number is determined from the whole population.

**A sample is a finite part of a statistical population whose properties are studied to gain information about the whole (Webster, 1985). When dealing with people, it can be defined as a set of respondents (people) selected from a larger population for the purpose of a survey.**

**Population**: It is a group of individual persons, objects, or items from which samples are taken for measurement. For example, a population of presidents or professors, books or students.

**Sampling** is the act, process, or technique of selecting a suitable sample, or a representative part of a population for the purpose of determining parameters or characteristics of the whole population.

A representative sampling plan ensures that the selected sample is sufficiently representative of the population

# Advantages of Sampling

Let us now understand the advantages of sampling in research.

Studying a small group of people would involve lesser resources as compared to the whole population. It would also save much time to study a small group rather than whole population. It has also been proved that the results obtained through a smaller group is more accurate than the results obtained through a census. Sampling also helps the researcher to have accessibility to the elements of the population.

## Universe of study

The prior step towards sampling is to decide the universe or population of the study. For example, undergraduate students of a city or a state, lactating mothers of a community, all child labourers in a state. Now we are clear that our units of sample will be from these well defined elements.

## Sampling design

Another most crucial part is to select the sampling design. This depends on the type of population and the objectives of our study.

## Sampling Frame

Sampling frame is the list of all those units belonging to the population from which the sample will be drawn. The difference between population and sample is that population is general whereas sampling frame is more defined and specific. Let us consider the above examples.

The sampling frame for these population can be undergrad students of government college, lactating mothers of a community registered with the Anganwadi or the health centre, all child labourers below the age of 11 years.

## Sample size

Next aspect is to determine the sample size which depends on the nature of characteristics of the units under study and their pattern of distribution under study. Here comes the chances of sampling error.

## Sampling error

It comprises of the differences or variation between the sample and the population, which is due to the selection of the units. In social sciences unlike physical sciences, human behavior is studied which tend to be unpredictable at times leading to an error. Therefore to reduce error it is required that the sample should be representative, adequate, unbiased, and objective.

# Types of sampling

## Probability Sampling

It is based on random selection of units from a population. In other words every unit in the population has equal chance or probability of being selected as a sample.

For example, all children in a classroom. Every child has equal chance of being selected as a sample. There are different types of probability sampling based on the population to be studied. They are

- Simple random sampling
- Stratified random sampling
- Systematic sampling
- Cluster sampling

### Simple Random Sampling

It is usually called the lottery method and is carried out with a homogenous group of population. In this every unit of the population has equal chance of being selected and the elements or units are chosen randomly. **Example**: students in a classroom, employees in an organisation etc.

### Systematic sampling

This type of sampling provides an even spread over the population leading to greater accuracy. In systematic sampling a fixed interval, say *n,* is chosen to draw the sample from population. In other words, if n=3 then every 3^{rd} element from the starting point is chosen in the sample. If the starting point is 2, then every 5^{th}, 8^{th}, 11^{th} ……element will be taken in the sample.

### Stratified Random Sampling

It is a method of sampling which involves division of the population into subcategories or subgroups based on similar characteristics. Units of the sample are chosen from these subgroup or strata. Stratification helps improve the representativeness of the sample. For eg. If we want to study students of high school, we may first divide the population on the criteria of different grades and gender of the students.

### Cluster Sampling

This type of sampling is done with large scale studies where the population is spread over a large area. This helps in minimizing cost as well as time for the study. For the study, the area is divided into smaller areas or ‘clusters’ and then stratified or simple random sampling is done in these clusters. Since this involves multiple stages of sampling hence also termed as ‘**multistage**’ sampling

## Non probability Sampling

**Non Probability **sampling is based on the subjective judgement of the researcher. Unlike probability sampling, it is a sampling technique where the probability of the members being selected for a sample cannot be determined. Selection of the units of the sample solely depends on the knowledge and experience of the researcher.

- Accidental/convenience sampling
- Purposive sampling
- Quota sampling
- Snowball sampling

### Accidental sampling

Also referred as convenience sampling, where the selection of the units from the population depends on how readily and conveniently they are available for the study. For eg. If our population is undergraduate students of a city. We may choose any college as per our convenience and draw the sample from them.

### Purposive sampling

This technique is used when the researcher has to make use of his/her judgment to identify the units which fulfil the purpose of the study. The researcher would select only those units which serve the objective of the study. The difference between convenience and purposive sample is that, in convenience sample the units are selected based on the proximity and ease of availability of the units of the population whereas in purposive sampling the focus is on the specific purpose rather than convenience or accessibility. In other words, units are selected based on their specific characteristics and not on their proximity. Even if the elements are difficult to access, they would be included in the sample if the researcher thinks that they serve the purpose of the research.

### Quota sampling

This type of sampling ensures inclusion of diverse elements of the population in the sample and ensures that exact proportions as in the population, are represented in the sample. For eg. Let us say we want to study pregnant and lactating mothers in a district. From the Anganwadi registers it is found that there are a total of 1000 women in this category of which 300 are lactating mothers, 400 pregnant women and 200 pregnant and lactating mothers. If we take a sample size of 100, then the sample must include at least 30 lactating women, 40 pregnant women, and 20 Pregnant and Lactating women.

### Snowball sampling

In this type of sampling the researcher starts with a few respondents who in turn guides towards other respondents. This type of sampling is done in a population where the elements are unknown, *viz*, drug addicts, HIV positive, pickpockets etc. who may be hesitant to identify themselves and can be accessed only with reference from other respondents.

# Conclusion

These were some of the sampling techniques applied in social work research. The choice of the sampling method depends on the requirements and considerations related to individual project. These may include the characteristics of the population under study, objective of the analysis, time available and of course resources available. After reading this article I hope you have got an understanding of the sampling in social work research.

### Sources

https://www.statisticshowto.com/probability-and-statistics/statistics-definitions/sampling-frame/