Sample Design
Sample
A sample or sample unit is an element or group of elements out of the total population on which the research study is based. The definition of a sample unit varies with the objectives of the sample survey because the objective of the sample survey decides what data to collect.
Example
- To study the job satisfaction level of 2000 employees working in an organization following are the sampling units.
Population: total employees working in an organization
Sampling unit: Any individual employee of an organization
Sample: The sample is the subset determined or decided for the study. It can be 100 employees or 200 employees. These numbers of employees are selected either randomly or with specific criteria.
Sampling Methods
The sampling method is the technique for selecting the required number of respondents to collect the primary data. The total population is the area of research from which the data will be collected to conduct the research. Sampling methods enable the researcher to collect necessary data for research. Sampling methods enhance the focus of research and avoid errors in the data. The wrong selection of sampling methods may be responsible for the collection of unnecessary data which leads to errors in research. The following are various methods utilized to conduct the research study.
Sampling methods can broadly classified into
Probability Sampling
- Simple random sampling
- Systematic sampling
- Stratified sampling
- Multistage Cluster sampling
Non-Probability Sampling
- Convenience sampling
- Quota sampling
- Judgement sampling
- Snowball sampling
Probability sampling
Probability sampling is the technique that allows researchers to select any sample from the total population of the study. The total population is the expected group of respondents from which the sample will be collected for primary data. In this technique, there is an equal chance of all the respondents being selected as a sample for the study. Probability sampling is implemented in four ways.
Simple Random Sampling
In random sampling methods, the samples are selected randomly without any criteria considering all the population is homogeneous and a small sample can represent the whole.
Examples
In manufacturing organizations such as manufacturing of MCB( Miniature Circuit Breaker), random sampling methods are used to determine the quality of finished products as well as semi-finished products. After producing a lot of hundred product units the selection of 8 to 16 units was done randomly to inspect whether the products were produced as per the expected standards. If any one unit help farts full lots get rejected. Before starting manufacturing the quality of input material or parts is also done by sampling methods to ensure the expected quality of supplied goods.
Systematic sampling
In the systematic sampling method or technique, the samples are selected based on specific intervals such as every bth element selected as a sample where b indicates the skip interval. This skip interval is calculated by following the formula
Skip interval (b) = sample size/population size
Examples
- Systematic sampling methods are used in the manufacturing organization to evaluate the quality of the products while manufacturing to know whether the quality is maintained in the manufacturing of the product consistently. It is essential to evaluate after a specific interval. Systematic sampling allows us to evaluate the manufacturing by selecting the products after specific intervals. This interval keeps track of the evaluation of the manufacturing of products as per expected quality standards.
- To study the consumer’s perspective or perception or behaviors in the mall systematic sampling can be useful. Systematic sampling methods allow the researcher to take the response of visitors in the mall after a specific interval in this way they represent the maximum portion of all the visitors.
Stratified Sampling
The stratified sampling technique involved dividing the whole population into distinct subgroups known as strata. Each sample unit of a stratum is homogeneous and representative of that group. The stratified sampling technique divides the whole population into strata and then selects members from each stratum proportionally. These techniques enable researchers to collect accurate and all types of samples from the whole population.
In this way, a variety of samples is collected by using the different types of data in which similar sample units are Classified. Stratified sampling provides data accurately and completely which leads to more accurate and reliable results than the random sampling method.
Examples
- In Education research stratified sampling was implemented as follows
Population:
Educated students
Strata:
Education level (UG, PG, M. Phil. Ph. D. SSC, HSC School level)
Explanation:
The above example shows the classification of the research population into subgroups or strata which includes homogeneous data. Then samples from each education level were selected as a representation of all the subgroups. In this way, all the varieties of samples were collected. Strategy by the sampling methods allows a researcher to collect the data from all types of Education levels starting from the school level to the Ph. D.
- Marketing research conducted by LED TV company
Population
LED TV users
Strata :
Operating system (Android TV, Google TV, Other)
Description
The above example shows the data collection by stratified sampling method in which classification of the population is done based on operating system users.
Cluster Sampling
Castle cluster sampling techniques involved dividing the world population into various clusters which consist of all the varieties of samples. After dividing the cluster few of the entire clusters were selected randomly. All the samples in the selected clusters are included in the study.
Cluster sampling is implemented in the case where the population size is very large and geographically spread along long distances. It helps to get the samples from various clusters that cover the largest part of the population in that region of population. Cluster sampling covers a Higher degree of variability which helps in collecting a variety of samples. A higher degree of variability enables researchers to address each type of issue or data required for study.
Cluster sampling in Agricultural research
cluster sampling technique implemented in agricultural research to collect the data as follows
Population
Farmers In a specific region or geographical area
Clusters
Farmers from various villages or communities in that selected region.
Description
In the above examples, the population size is all the farmers of a specific region. Several villages are there in specific regions which are large and geographically separated also. To address the variety of the population or samples it is essential to select major groups known as clusters. The clusters are formed according to the number of villages or farming communities left in that region so every village is entirely a cluster that will be selected as a sample. Out of several clusters, a few of the clusters entire clusters were selected for the study.
Non-probability sampling