Sampling Design
1) Introduction
Research is based on facts. Since the findings of the research can be drawn from the analysis of all types of data obtained from the research, information has to be collected through various methods of gathering facts. Due to the resources, means, time, and other practical complexities involved in researching the Universe or all the units of research in the research work, it is necessary to select and study only a few representative units from the Population Unit. The census method cannot be considered suitable even if not every unit can be deduced from the research. Therefore, in such a case, a certain representative unit is selected from the association. This is why the process of filling the sample and sample selection is called sampling. In social research, the method and process of selecting a certain representative unit from the totality of research and applying its findings in totality are called sample selection.
The first formal and systematic adoption of the sample selection process in social research was made by the US Department of Statistics in 1940. Previously, sample selection was used only in the field of social research. Initially, this method was used in the field of study and research in the field of agriculture besides social research, but today the practice of sample selection in the field of research in natural and social sciences including physics, zoology, sociology, population, agriculture, economics, sociology management is very practical and theoretical.
In general, when collecting data, facts can be collected in research from two basic methods. The first method is Census and the second method is Sampling. Each unit involved in the research is researched through computational methods. Information related to each unit is collected and a conclusion is reached. For example, the census of Nepal every ten years is similar. This computational method does not exclude a single unit of research. In this method, all the necessary information can be obtained by taking data related to each unit. This does not raise any doubts about the actual situation. If the researcher is independent and the correct measurement method is adopted.
The census method cannot be considered practical in cases where the field of research is large and cannot cover all the units. Due to the fact that research requires more resources, tools, and time than the computing method, it is found that the sampling method has been adopted everywhere in the research rather than the computing method. Failure to choose the right and representative model in this method will undoubtedly reduce its credibility. The sample selection method assumes that the smaller unit represents the whole. In a very small research area, it is better not to follow the sample selection method.
Different scholars have their own arguments regarding the sampling method. “Sampling is taking of any portion of a population or universe as representative of that population or universe,” Kerlinger said. In this definition, he considers the choice of a unit that can represent totality as a model. In which some representative part or unit is selected from the totality.
Gay’s idea is as follows: “A population as a group to which a researcher would like the results to be generalizable” The sample selection was taken in this definition emphasizes that the whole should be generalized. The definition of Goode & Hatt is very clear and concise, according to him - “A sample, as the same implies, is a smaller representation of the larger whole.” Is taken as
Sample selection is used in everyday practical life in addition to research work. For example, to determine the quality of rice in a bag based on the grains of There rice taken from a bar. To determine the blood group of the body from a drop of blood. Determine whether a piece of vegetable is ripe or not. Kasaudi is taken by touching a few grains of rice. The process of sample selection can also be understood from the process of drawing the conclusion of the whole rice. What is clear from this is that sample selection is a process in which some representative units are selected from the entire study unit. The selected units are considered to represent the totality.
Sampling Design is also a systematic and step-by-step process. Therefore, the researcher has to be careful that the units selected in the sample need to be representative. For this, the Universe must be well defined and deciphered. Sample selected unit. This sample selection pattern has no meaning unless it is representative. In this regard, Kothari said that the sample survey format is a definite plan to obtain a sample from a given population. (A sample design is a definite plan for obtaining a sample from a given population)”.
In conclusion, what can be said about sample selection is that the act of selecting a certain representative unit from the total population is called sampling, and those representative units are the sample that represents the totality. The use of this sample selection method seems to be very practical as the census method is more expensive in terms of time, resources, and means. The characteristics of the sample selection structure based on various analyzes can be mentioned as follows.
2) Steps of Sampling
The sample selection method has to be done in a systematic, phased, and procedural manner. There is a situation where the selected sample unit has to be made representative and while choosing another sample one has to be equally careful not to use bias, error, and impractical processes. In some cases, a special sample selection method has to be used considering the characteristics of the units in the Universe and the position of the boundaries in them. In this case, sample selection is a plan and a strategy. Therefore, Sampling Design should be prepared in a phased manner. The steps are as follows.
a) Identification of the Nature of the Universe
What is the nature of totality, that totality, individual, organization, or community? That must be done first. A group of individuals and a group of organizations can exist as a whole. Which is called population. Which consists of several units. Therefore, it should be clear about who and where is a group or organization, who is an individual, and where. We have to consider its nature.
b) Identification of the types of Population,
There are three conditions for population type. In which the totality is certain or uncertain? What is the target population? And what is the fixed population? The results of which help to choose the type and type of sample selection process to follow.
c) Determine of Proposed Sampling Unit and Size
What is the sample size to be selected? The percentage or number to be chosen as a sample should be determined on the basis of the person, place, institution, etc. to be selected.
d) Listing the Units
All units in totality or certain units should be listed based on their characteristics. In this way, the list is prepared by writing the name of the unit or listing the unit by any sign.
e) Identification of the Sampling Design
What type of sample selection should be selected and what are the procedures? That has to be decided. The method should be based on the purpose, time, seven, nature of the instruments, and units of your study.
3) Sampling Error
If there is an error in the sample selection, the result cannot be generalized. In the conclusion of the research, the difference between the true value and the parameter value of mathematics is called an error in sample selection. In this regard, Lindquist states that the difference between the estimated mathematical sum (sample) and the constant mathematical sum (population) brought from the sample can be called an error in the sample selection. A sample error is a difference between a parameter and an estimate of that parameter which is derived from a sample
Let’s look at the fact of error in sample selection as an example - as a researcher determined 95 mediums from 20 units of his sample which have a totality of 100 and calculated 10 means from the totality of 100 units, there is a difference of 0.5. Which can be called a sampling error in sample selection. Therefore, errors in sample selection do not arise solely from the researcher. Basically, two conditions play a role in this type of error. Like-
a) Variability in the population
Once you start doing research, there is instability in the totality and in the trend. As a result, the mean value of the sample cannot completely equal the mean value of the whole. When errors appear (there is a difference between a population value and sample value).
b) Size of the Sample
There is an error in the selection of the sample even when the sample could not be selected as an adequate sample study unit.
Sample selection errors can be divided into four parts.
1. Sampling Errors
2. Sampling Biases
3. Non-Sampling Errors
4. Non-sampling biases
Error in sample selection refers to the error that occurs when a suitable, adequate, reliable representative sample cannot be selected, while bias in sample selection refers to the same non-random method and bias in sample selection due to insufficient numbers. Errors other than sampling are errors that may be present in observations and interviews, errors in statistical analysis and measurement, and errors in respondents. Similarly, errors other than sample selection refer to errors arising from biases in observation and measurement. The variable error refers to the error received from the sample. This arises from the errors and biases in the sample selection.
4) Precautions to be taken while choosing the sample
Sample selection is a complex and sensitive task. In the case where the selected samples do not represent completeness, the whole research is justified. Conclusions from the research of incorrect samples cannot be generalized. The following precautions have to be taken while selecting a sample for research work as it is not possible to select a suitable sample and the research itself will become unproductive and unreliable due to the expenditure of resources and time.
a) Analysis of Homogeneous and Heterogeneous
When choosing a sample, it is necessary to have a good knowledge of the universe. How many ordinary and how many extraordinary units are there in totality? On what grounds are each unit similar or not? Researchers should be well aware of this. You must have a separate list for each unit.
b) Selection of Representative Units
The choice of the representative unit should be taken into consideration whether there is uniformity in the units of the sample taken for research.
c) Accuracy
When choosing a sample, the researcher should choose the sample according to the scientific method without feeling biased.
d) Selection of Independent Unit
Each unit of sample selection should be independent. In addition, the scope and boundaries of research should be clearly defined. Depending on the topic of the research, different sample selection methods can be adopted as required.
e) Pilot Survey
When choosing a model, the economic status of any society, social diversity, gender diversity, regional diversity, etc. should be included. Therefore, it is necessary to conduct a pilot survey not only on the basis of the second fact but also before selecting the sample in the relevant area.
f) Determine Adequate Size
When selecting a sample, the sample can be selected in a lower proportion from the larger and homogeneous units, while the group with different characteristics, which is smaller, should be selected in a relatively higher proportion. Sample selection needs to be representative and adequate. It is generally advisable to take more than 15 percent when determining the size of the sample selection. More sample units should be selected for larger integrity and less for smaller integrity.
1 When choosing a sample, you should coordinate your time, resources, and means to some extent. Samples should be selected with as few sampling errors as possible.
2. The sample should be selected in such a way that the information to be collected is clear and in line with the objective.