Site icon The Skull Session

How to design a sample as part of research design?

Featured Image for HRP

When a student of BHM, has to do a research as part of their syllabus in the final year; one of the daunting task is to get a through understanding on the concepts of designing the research. This articles aims to be useful in that regard.

What is sample design ?

In the world of research the term design refers to a plan of work, thus sample design refers to the plan regarding the sample that will be selected from the population for the purpose of study.

Why is sample design needed?

When a research is undertaken, one of the key factors to be considered is “who” are the respondents. In other words, respondents are the source for collecting the primary data or raw data. Practically, it is impossible to collect data from all the respondents (units) from the field of study (universe/ population); hence we have to select few respondents that will be an epitome to the entire population.

When we do this, we are generalising the characteristics of the population based on the response from few units that we have selected. It may be noted at this point that the sample collected should be true representation of the population for this method to be effective. If the population is homogeneous in nature samping will be very effective.

Consider this hypothetical situation A batch of cake is made from the same batter mix. We can take a piece of cake to understand or estimate the quality of the entire batch of cake. Here, the universe is homogeneous in nature (cake batter mix) and the sample (piece of cake) is a true representation of the entire cake.

However, in case of heterogeneous population the possibility of getting an apt sample that will be a representation of the population will require some careful planning. Thus a sample design is necessary for conducting a valid research. Moreover if the population on universe is infinite in nature, sampling is the only method of data collection.

Consider this hypothetical situation Scenario – I : A study is conducted to understand the degree to which the students have understood the subject of research methodology. Here if the smartest students are taken as sample, we shall get positive results. however if the weakest students are taken as sample, we shall get negative results; none of the samples are the true representation of the fact. Scenario – II : The same study is conducted for the entire students who have enrolled for BHM across India, now the universe/ population is huge hence can be termed infinite in nature. In both the scenarios mentioned, it is imperative to select few units as the representation of the population in such as way that there is maximum accuracy.

What is population/ Universe?

Population or universe refers to all the units under study. In the illustration there are 32 units under study, hence the size of the universe is 32; that is N= 32.

The illustration is of a heterogeneous population, consisting of four categories namely boys, girls, women, & men.

What is sampling?

Sampling is the process of selecting few of the units from the universe for the purpose of the study.

In this illustration we have sampled one from each category. Thus from a universe of 32 units, we have chosen 4 units for study; Hence the sample size is 4.

Which are the types of sample?

Sampling techniques in research

Key differences between the sampling techniques:

Non-Probability samplingProbability Sampling
The units are deliberately chosen by the researcher.The units are randomly chosen
Personal element is present in the samplingNo personal element is present in the sampling
High risk of bias Low risk of bias, as every unit in the universe has an equal chance of inclusion
Sampling error cannot be estimatedSampling errors can be estimated 
Not suitable for large inquiriesIt is suitable for large inquiries
Less expensive method and can be used in smaller inquiries.Various mode of sampling are available and can be used in large scale inquiries
Non probability vs Probability sampling

Types of Non probability sampling

Convenience sampling

Convenience sampling – Primary data is collected from individuals who are most accessible to the researcher. This method cannot create generalisable results though.

Eg. you have asked your friends in your batch to complete the survey.

In the illustration, man depicted in yellow is the researcher. In case of convenience sampling, very few people from the population is only considered for study, refer the illustration for representation.

Voluntary response sampling

Voluntary response sampling – It is similar to convenience sampling, here the survey is sent out to all the people in the college, and anyone can volunteer to respond.

E.g. You create an online survey and share in every batch group via whatsapp

Voluntary response sampling has further reach, since an online form such as google form, survey monkey, microsoft form and such devices may be used.

Purposive sampling

Purposive sampling – Here the researcher uses their judgment to choose the sample which is most appropriate for the purpose of the research.

E.g. you choose to conduct a survey with specific people such as hoteliers, restaurants, manufacturing units and so forth.

In the illustration, only the women are selected as sample from the population.

Snowball sampling

Snowball sampling – If the population is difficult to access, snowball sampling can be used. Here a participant sends the survey to other people (friends, family ect), and asks them to forward the survey to people known to them; thus helping to gather more data.

From the illustration it can be seen that the researcher sends an online form to two respondents and asks them to forward the same to people known to them; and convey the same message. As a result comparatively deeper reach can be achieved.

Types of probability sampling

Simple Random Sampling

Simple Random sampling is the purest form of probability sampling. Each member of the population has an equal and known chance of being selected. When there are very large populations, it is often difficult or impossible to identify every member of the population, so the pool of available subjects becomes biased.

Simple random sampling refers to any sampling method that has the following properties.

If all possible samples of n objects are equally likely to occur, the sampling method is called simple random sampling.

An important benefit of simple random sampling is that it allows researchers to use statistical methods to analyze sample results. For example, given a simple random sample, researchers can use statistical methods to define a confidence interval around the sample mean. Statistical analysis is not appropriate when non-random sampling methods are used.

In simple random sampling, each unit has equal chance of being selected.

There is no criteria for selecting the units.

Statical methods can be applied for this type of sampling.

There are many ways to obtain a simple random sample. One way would be the lottery method. Each of the N population members is assigned a unique number. The numbers are placed in a bowl and thoroughly mixed. Then, a blind-folded researcher selects n numbers. Population members having the selected numbers are included in the sample.

Example: We have to select 100 student from the college as sample. The total number of students is 480. We can assign each student a number ranging from 1-480 and randomly select any 100 numbers.

Complex Random Sampling

Systematic sampling is often used instead of random sampling. It is also called an Nth name selection technique. After the required sample size has been calculated, every Nth record is selected from a list of population members.

As long as the list does not contain any hidden order, this sampling method is as good as the random sampling method. It’s only advantage over the random sampling technique is simplicity. Systematic sampling is frequently used to select a specified number of records from a computer file.

Example: We have to select 100 student from the college as sample. The total number of students is 480. We can assign each student a number ranging from 1-480 and decide to select every 3rd number. Hence, the final sample size will be 160 (480/3)of which the first 100 may be chosen.

Stratified sampling is commonly used probability method that is superior to random sampling because it reduces sampling error. A stratum is a subset of the population that shares at least one common characteristic. Examples of stratums might be males and females, or managers and non-managers. The researcher first identifies the relevant stratums and their actual representation in the population.

Random sampling is then used to select a sufficient number of subjects from each stratum. “Sufficient” refers to a sample size large enough for us to be reasonably confident that the stratum represents the population. Stratified sampling is often used when one or more of the stratums in the population have a low incidence relative to the other stratums.

Stratified sampling refers to a type of sampling method . With stratified sampling, the researcher divides the population into separate groups, called strata. Then, a probability sample (often a simple random sample ) is drawn from each group.

Stratified sampling has several advantages over simple random sampling. For example, using stratified sampling, it may be possible to reduce the sample size required to achieve a given precision. Or it may be possible to increase the precision with the same sample size.

Example: We have to select 100 student from the college as sample. The total number of students is 480, of which 240 boys and girls are present. The sample should contain equal proportion of boys and girls; hence the first 240 numbers can be boys of which 50 will be selected as sample & rest will be girls.

Cluster Sampling

Cluster sampling refers to a type of sampling method . With cluster sampling, the researcher divides the population into separate groups, called clusters. Then, a simple random sample of clusters is selected from the population. The researcher conducts his analysis on data from the sampled clusters.

Compared to simple random sampling and stratified sampling , cluster sampling has advantages and disadvantages. For example, given equal sample sizes, cluster sampling usually provides less precision than either simple random sampling or stratified sampling. On the other hand, if travel costs between clusters are high, cluster sampling may be more cost-effective than the other methods.

Example: MG University has over 223 colleges affiliated to it. Since, it is spread out across various districts of kerala, it may not be feasible to collect data from all the college, hence it may be decided to collect sample from only professional degree courses under the university; hence 10 colleges are selected as clusters.

Multi stage sampling 

 This is a further development of the idea of cluster sampling. This technique is meant for big enquiries extending to a considerably large geographical area like an entire country . Here, the first stage may be to select large primary sampling units such as states, then districts, then towns, and finally certain families within towns. If the technique of random sampling is applied at all stages, the sampling procedure is described as multi stage random sampling.

Characteristics of good sample design

Thus the sample design is an important aspect of the research design process. A good sample design should have the following characteristics:

Skip to toolbar