Research: Population and Sample
Research: Population and Sample

Research: Population and Sample

Lead Author(s): Dr. Rafeedalie

Source: Edmodo

Student Price: FREE

In this homework assignment students will be asked to understand population, sample and various sampling techniques 

Population and Sample


Reading this unit, the student will able to

1. Comprehend the concept like, Population, Sample, sampling, sampling error
2. Understand the various sampling techniques of Random Sampling and Non
Random Sampling theories

​1. Population

 In research terminology the Population can be explain as a comprehensive group of individuals, institutions, objects and so forth with have a common characteristics that are the interest of a researcher. The common characteristics of the groups distinguish them from other individual, institutions, objects and so forth. The term universe is also used as synonyms to population. Suppose a researcher proposed to conduct a study on awareness and use of ICT among the secondary school teachers in Telungana, the entire secondary school teaching community in Telungana constitutes as the population of the study. 

Sometimes population can be counted easily, which is called finite population. Population of medical students is an example of finite population. The unlimited or unknown number of population can be called as infinite population. The adolescents, youths in Telungana can be treated as examples for infinite population, though they can be counted but in complex procedure. 

Any value which is identified or measured from the characteristics of entire population can be called as Parameter. The process of conducting a survey to collect data from the entire population is called a census.

​2. Sample 

In social science and educational research, practically it is not possible to a researcher to approach all the individuals\elements in a population for the purpose of data collection. Instead they select and approach a representative group of individuals/elements who falls under the particular population to collect needed information regarding the group. Based on the results, the researcher generalizes the characteristics of the representative group as the characteristics of population. This 2 small group or representative group from a population is called as sample. So sample can be defined as the small portion of a population selected for a particular study. The sample should clearly represent the characteristics of intended group. According to Young “A statistical sample is a miniature picture of cross selection of the entire group or aggregate from which the sample is taken”. The process of conducting a survey to collect data from a sample is called sample survey. The value which is identified or measured from the characteristics of the sample can be termed as statistic.

​3. Sampling 

The process of selection or the drawing of the accurate representation of a unit, group or sample from a population of interest is called as sampling. Sampling can be done through various sampling techniques in accordance with the nature of the sample as well as the subject matter of the study. It is the Sampling procedure, which will decide the accurate representation of the sample selected for the study as well as the relevance of generalization made from the research. 

4. Sampling Unit 

Each individual or case that constitutes a sample is called a sampling unit or sampling element. For example if a sample constitutes 200 teachers, each teachers in the sample are considered as a sampling unit. 

5. Sampling Frame 

Sampling frame is the list of subjects/people under the study, such as household,, students, teachers, principals and so forth. The list should be comprehensive as well as latest. For example, Telephone directory, Students data base from department of school education, list of school principal from the official website of concern department and so forth.

 6. Sample Size 

As name indicates sample size is the total number of sample selected for the study. For example, it is the number of teachers, students or stakeholders from a researcher intended to collect information regarding his research questions. There is no notion about the minimum or maximum number of sample; instead the sample size should be optimum. Usually the sample size is denoted by the letter (n).

​7. Sampling Error 

The variation between the means of sample groups as well as population mean is called sampling error. It can be understood through the following example. A researcher planned to conduct a study on Emotional Intelligence of secondary school 3 students in Telungana state. Definitely the researcher has to selects accurate representation or optimum sample from the large population of his study. Suppose the researcher has selected ten groups or samples each consisting of 200 students from same population. He administered his research tool in each sample, collected the data, organized, scored and found the mean scores of each group. Finally he can see that each group shows differences in their mean scores with another group or sample as well as with the population mean. This is because of; a random sample will not be identical representation of a population. A few would be relatively high, a few relatively low, but most could tend to cluster around the population means. This variation of sample means is due to sampling error. The term does not suggest any mistake in the sampling process, but merely describe the chance variations that are inevitable when a number of randomly selected sample means are computed. Hence the variation between the sample mean and the population mean are called sampling error.

​8. Representative and Biased sample 

Representative samples are the samples which are closely match the actual characteristics of the population from where the samples have been drawn. When a researcher select the sample through systematic and scientific way and ensure the optimum sample size, he/she can ensure the representative sample for his/her study. Biased sample can be defined as the sample which is not representative of the actual/common characteristic of the population from which it was drawn. A researcher may select biased sample intentionally or unintentionally. For example when a researcher intents to establish a favourable outcome over others, he may adopt biased sampling technique to ensure the indented results. Suppose if a researcher want to prove relationship between the Intelligence and school discipline, he may select the students as a sample for the study from the class who maintain high discipline as well as high intelligence, where as there might have several classes in that particular school where people are with high IQ but low discipline. In unintentional cases the same thing might be happen through the random selection of the particular class from a several classes of the school. Even though it is an unintentional selection of the sample, it should have affected the result of the study as it was not the real representation of the actual characteristics of the population.

​9. Sampling Techniques

 Sample can be selected through different methods. Blalock (1960) classified the sampling methods in to two categories on the basis of the nature of selection of the sample units. They are given below. 

  • I) Non-Random sampling techniques (Non- Probability Sampling) 
  • II) Random sampling techniques (Probability Sampling) 

A. Non-Random sampling techniques (Non- Probability Sampling) 

Non random sampling techniques are the techniques in which the researchers select the samples from the population without randomization. Here the samples might have selected at the discretion of the researcher. In this sampling there is no means of judging the probability of the element or group of elements, of population being included in the sample. Important non random sampling techniques are given below.

​I) Convenience 

Sampling When the researcher selects sample for the study at his own convenience is called as convenience sampling. For example an investigator who is doing research on the topic of social skills of adolescence and he may take students of X class as sample for his study, because he has been the class teacher of the same class and happens to be friendly with the class. This is what is called as convenience sampling. Such samples are easily available and economical but it makes systematic errors and may leads to false generalizations. Convenience sampling is also called as haphazard as well as accidental sampling.

 II) Quota Sampling

 In this sampling the investigator initially sets some relevant categories of people and decides the number of units should be selected for the study as a sample. Such as male= 10, female=10; or science students=20and humanities students=20 and so forth. Quota sampling has some benefit over the convenience sampling because it ensures some differences or inclusion of variety of elements in the sample. But the problem is that here the researcher select the categorized people at his/her convenience. There for select those who are easiest to interview or administer questionnaire, so sampling bias can be take place. Although it has some limitation it enables the investigator to introduce a little control over the sample.

​III) Purposive Sampling 

It is also known as judgment sampling. It is valuable in special circumstances. Judgment sampling is used in exploratory research or in field research. The researcher may exercise his own judgment or uses the judgment of an expert in selecting cases. In purposive sampling the researcher never knows whether the cases, selected represent the population. Purposive sampling is suitable to select unique cases when the researcher knew that they might be providing relevant and valuable information that he or she requires.

 For example a researcher wants to study the aggressive attitude of children with anti social behavior. It is very difficult to list all children with anti social behavior from the list. Here the researcher may use different methods to identify the cases and approach them to get relevant data. The prime concern of judgment sampling is that to understand or judge the researcher that who can pour the accurate information regarding the topic of the study to the researcher. 

Judgment sampling is economical, more convenient, easily accessible and select only those persons who can give relevant information to the research area. The main limitation of the purposive sampling is that it does not ensure the actual representation of the selected sample of the population instead it concentrate only the ability of the sample to pour relevant information regarding the topic of the study.

​IV) Snowball Sampling 

It is a sociometric sampling method and also known as network, chain referral or reputation sampling method. In this method the researcher starts collection of data from the person who known to the researcher. At the end of the data collection the respondent will be asked to provide the contact information of another respondent who can give relevant information regarding this area of the study. These processes are repeated and get more respondents and relevant information to the researcher. Snow ball sampling is more useful when there are small possibilities to get the information regarding the population or the population is unknown.

​B. Random sampling techniques (Probability Sampling) 

Random sampling methods are the methods which ensure the probability of each element in the population for being selected as sample unit for the study. Here the sample units are not selected at the discretion of the researcher instead it follows certain procedures which ensure the probability of each unit in the population of 6 being included in the sample. Hence these methods are also called as Probability sampling methods. Random sampling is free of bias in selecting sampling unit. Major random sampling methods are following.

​I) Simple random sampling

 It is the simplest form of random sampling. In this sampling technique each elements of population might have given equal chance to be selected for the study. Randomness completely depends on the procedure of selection of sampling units from the population.

Basic requirements of simple random sampling

1. Prepare a comprehensive list of all the units in a population of interest
2. Design a method where all the units get equal chances to be selected as a
3. Ensure a systematic process of selection where one unit of selection has no
impact on the chances of selecting another unit.

​The best method that can be used for simple random sampling is lottery method. For example if a researcher want to select 20 students from a class which consists of 100 students. He/she can write the names or roll numbers of the whole students on separate slips of paper in equal size and colour- and fold them in similar way. After that the whole slips should be placed in a box and shuffle thoroughly. An out person may be invited to pick twenty slips from the box as he wish. The selected students (slips) are considered as the sample for the study. Here all the 100 students have got equal chances to be selected. 

Advantages of simple random sampling

  • i. Every person has an equal chance of being selected
  • ii. It follows a systematic procedure for sample selection
  • iii. It serves as a foundation of all other random sampling techniques
  • iv. It is suitable when the population is relatively small; sampling frame is
  • comprehensive and up to date
  • v. As the sample size increases, it becomes more representative of universe.
  • vi. It is economical as well as yield accurate result for the study

Limitation of simple random sampling

  • i. Practical difficulties to prepare a comprehensive list of population
  • ii. Updating population is big task
  • iii. Large sample size is required to establish the reliability.
  • iv. As the population widely scattered, it becomes costly as well as time
  • consuming
  • v. If there are more heterogeneity among the unit of population, a simple random
  • sample may not necessarily represent the true characteristics of population
  • vi. Unskilled and untrained researcher may cause for making wrong
  • generalization.

​II) Stratified random sampling technique 

When the researcher needs stratification of population based on single characteristics or attributes such as male and female, urban and rural, married and unmarried and so forth he/ she warranted the stratified random sampling technique. Here the population is divided in to two or more strata. For example, if researcher want to study the emotional intelligence of graduate students. He can stratify the population in to three such as science graduate, social science graduate, commerce graduate. These categorized populations are called subpopulations. The usual stratification factors are age, sex, socio economic status, educational qualifications, locale, occupation, religion, cast, intelligence and so forth.

Advantages of stratified random sampling

  • i. It increase the precision in estimating the attributes of the whole population
  • ii. It provide more convenience in sampling
  • iii. Ensure the accommodation of the whole relevant strata of the population
  • iv. More representative of the population as it includes the each subgroup of
  • population
  • v. Free from bias to a great extend
  • vi. Through proper planning it can be economical as well as make timely

Limitations of stratified random sampling

  • i. Improper stratification may cause wrong results
  • ii. More strata requires large sample size
  • iii. Lack of proper planning may lead to too costly and more time
  • iv. Trained investigators are required for stratification

​III) Cluster Sampling Cluster sampling 

is a variation of simple random sampling. It is used when the population of the study is infinite and the population units are scattered across the wide geographical area. For example government of India wants to conduct a survey on the people attitude towards the Swatch Bharath programme. It is neither feasible to conduct a survey on all citizens throughout India nor justifiable to administer a questionnaire or conduct interview among any particular part of India. Instead in this type of study the researcher can use cluster sampling. In above stated problem the 8 government can select the sample randomly in multi-stage. Initially, government can select any 10 states from different parts of the country. Then from each selected state 4 districts may be selected and from each district 100 citizens may be approached for data collection. This sampling technique can be also called as area or multi stage sampling.

​IV) Systematic Sampling 

Systematic sampling can be defined as selecting or drawing of every nth item or person from a pre determined list. Such as selection of every 10th person from a telephone directory or every 6th person from a college admission register. For example if a sample of 250 were to be selected from a telephone directory with 2, 00,000 listings, one would select the first name by randomly from a randomly selected page. Then every 987th name would be selected until the sample of 250 being selected. If the last page were reached before completing the proposed sample size, the count would continue from the first page of the directory until it complete its intended sample size.


Distinguish between Population and Sample


Define the term Sampling


Differentiate between sampling frame and sampling unit with example.


What do you ,meant by sampling error


Write a short note on quota sampling


Describe the advantage and limitations of stratified random sampling


Dr. RAFEEDALI.E, Assistant Professor,
MANUU, CTE, Srinagar, 9419035681,