Types of Sampling Methods | Random Sampling, Non-Random Sampling (2024)

● It is also known as probability sampling or representative sampling.

● There is no room for discrimination in random sampling.

1. Simple or unrestricted random sampling
2. Restricted random sampling

(Unrestricted random sampling)

● This method is also known as unrestricted random sampling.

● The process used decides the chances of selection of an item, not an investigator.

● Under this type of random sampling, the samples are selected by using the following two methods:

1. Lottery method
2. Table of random numbers

● It involves the personal attention of the investigator while selecting a sample.

● It is not purely random.

● Important methods under this category are as follows:

i. Stratified random sampling

ii. Systematic sampling

iii. Cluster or multistage sampling

sampling.

(a) Stratified random sampling

(b) Systematic sampling

(c) Cluster or multistage sampling

● Example: To select two monitors randomly in a class of 40 students. First of all students are divided into two hom*ogeneous groups, i.e., boys and girls and then each one is selected from them randomly.

Merits

1. The sample taken is more representative of the universe.
2. It is easier to organise and administer because the universe is subdivided.
3. It ensures greater accuracy because each group contains uniform items.

Demerits

1. It is not possible if information about the population or ‘strata’ is not available.
2. If stratification is not done properly, the purpose will not be served.

● Under this method, the whole population is arranged ‘alphabetically’, ‘geographically’, ‘numerically’, or in some other systematic order.

● Then every ‘nth’ item is selected as a sample item. Where ‘n’ stands for any number.

● Like, every even or odd item.

● For better results, a list of items should be completely random and the first items should be selected using a simple random sampling method.

Merits

1. It is a very simple method and generally, the results are satisfactory.
2. Re-checking can be done quickly.
3. It requires the same amount of time and effort.

Demerits

1. It is possible only if the complete list of items is available.
2. It is feasible only if the units are systematically arranged.
3. There are chances of biasness.

Or

Multistage sampling

● It is carried out in multiple stages say, two, three, or four stages.

● In the first stage: The universe is divided into many clusters from which certain clusters are selected at random as the first-stage samples.

● In the second stage: The selected first stage samples are again subdivided into some clusters from which again certain clusters are selected at random as the second-stage samples.

● In the third stage: The selected second stage samples are again subdivided into some clusters from which certain clusters are again selected at random as the third-stage samples.

● The process of division and subdivision of clusters and selection of multistage samples is carried out until the sample size is reduced to a reasonable extent.

Merits

1. It is very helpful in large scale surveys.
2. It represents the population with reasonable accuracy.
3. It saves time and money.

Demerits

1. The division of population into clusters and sub-clusters is quite a difficult task.
2. The investigator needs to have detailed knowledge about the universe expertise in division and selection of clusters.

● The investigator selects samples on the basis of convenience or his judgment rather than on the basis of probability.

● On the basis of his own choice, he tries to select the best representative of the whole population.

● It is also known as purposive and deliberate sampling.

● Example:

● If a music teacher has to select five students from his/her school for participation in an inter-school competition. He/She cannot use a random sampling method.

● In this case, he/she will use his/her own judgment to select those five students from a big lot.

Merits

1. It is useful where the personal judgment of the investigator is important.
2. It helps in drawing the sample of a small size.
3. It helps in observing some characteristics in detail.

Demerits

1. It is not based on probability, it does not guarantee accuracy.
2. The selection of items may be affected by personal bias or prejudice.

● However, within the given quota, the selection of sample units depends upon the personal judgment of the investigator. So, this is a type of judgment sampling only.

● Example:

● In a survey of Reliance Jio network users, the interviewers may be told to interview 100 people living in a certain area.

● Out of those 100, 60% of the interviewed are to be working people, 30% should be students, and others to be 10%.

● Within these quotas, the interviewer is free to select the people to be interviewed.

Merits

1. It provides satisfactory results if quotas are allocated objectively.
2. Each part of the population gets representation.
3. Satisfactory results are expected.

Demerits

1. This method is subjected to personal bias.
2. It proves useful only if the interviewers are properly trained.

● Example:

● To estimate the average height of an Indian, the investigator (belonging to Delhi) can take a convenience sample from Delhi only and estimate the average height of an Indian.

● This method of selecting the sample is also known as ‘chunk’.

Merits

1. It is useful when the universe is not properly defined.
2. It is useful for the economy of time, money, and efforts.

Demerits

1. The sample items may not truly represent the universe.
2. The results obtained are often less reliable.

● In other words, larger the size of the sample, more accurate the results are likely to be.

● It refers to the differences between the sample estimate and the actual value of the characteristics of the population.

● Sampling errors can be of two types.

● Biased error: An error that arises on account of some biases or imbalances on the part of the investigators, informants, or instruments of counting, measuring, or experimenting is known as a biased error.

● Unbiased error: An error that does not take place on account of any bias with anybody but occurs accidentally may be due to a chance or due to an arithmetic error is known as an unbiased error. Such errors arise automatically without any motive.

● The magnitude of sampling error can be reduced by taking a larger sample.

● These errors occur in acquiring, recording, or tabulating statistical data.

● These are more serious than sampling errors because a sampling error can be minimised by taking a larger sample.

● However, a non-sampling error cannot be minimised even by taking a larger sample.