Wednesday, November 4, 2009

TYPES OF SAMPLING

There are basically two types of sampling:

· Probability
· Non-probability

Probability sampling: Is one in which every unit of the population has an equal probability of being selected for the sample. This remains the primary method for selecting large, representative samples for social science and business researches.

Advantages:

· High degree of representativeness
Disadvantages:

· The method is expensive and time consuming.
· Relatively complicated since it requires a large sample size and units selected are usually largely scattered.

Probability sampling is divided into the following different types:

Simple random
Stratified random
Systematic (interval)
Cluster
Multi-stage

1. Simple random: In this sampling the sampling units are selected randomly by one of the number of methods given below:

Lottery
Picking blind folded
Tippet’s table method or random numbers method
By first letter

Advantages:
Simplest and easiest to conduct
Sampling error is less
The researcher does not need to know about the exact composition of the population.

Disadvantages:

Researcher’s knowledge on population is not used.
Not possible if the researcher wants to break it in different sub-groups.
Produces greater errors in results than other methods.




2. Stratified random: When the population is divided into different strata or sub-groups and sample units are taken in a simple random method from each group. The final sample contains sample units from all these strata.

Stratified random sampling is of two types:

Proportionate
Disproportionate

Proportionate: When the sample units are chosen proportionate to the size of the strata.

Disproportionate: When it is not proportionate to the size of th strata.

Advantages of stratified sampling:

1.All groups of the population can be proportionately represented.
2. Comparison can be made among different sub-categories.
3. More precise than simple random sampling.

Disadvantages of stratified sampling:

. More efforts required than simple random sampling.
. The population has to be appreciably large to get statistically meaningful results.

3. Systematic (interval): The sampling is obtained by collecting of elements by drawing every nth person from a pre-determined list of persons. It is randomly selecting the first respondent and then every nth person after that. The number ‘n’ is called the sampling interval.

Advantages:

. It is easy, simple to use and a rapid method.
. Mistakes in drawing elements are relatively unimportant.

Disadvantages:

. It ignores all persons between two nth numbers paving the way of over/under representation of several groups.
. As each element has no chance of being selected, it is often not considered probability sampling by some.

4. Cluster: This sampling implies dividing population into clusters and drawing random sample either from all clusters or selected clusters.
Initial clusters are called primary sampling units; clusters within the secondary clusters are called multi-stage clusters. For example, dividing one city into various wards, each ward into areas, each area into neighborhood and each neighborhood into lanes and so on.

Advantages:

. Much easier when large population or large geographical area is studied.
. Less costly
. Respondents can be easily replaced.
. Characteristics of clusters can be estimated.
. Administratively simple to handle.
. Handy when it is inconvenient or unethical to randomly select individuals.

Disadvantages:

. Each cluster may not be of equal size, so comparison would no be on equal basis.
. Greater sampling error.
. Lacks representation

5. Multi-stage: In this method, sampling is selected in various stages but only the last sample of subjects is studied.

For example: For studying the panchayat system in villages, India is divided into zones = North, South, East and West. One state is selected from each zone say Punjab, Tamil Nadu, Assam and Gujrat. One district will be selected from each state. One block is selected from each district, and three villages are selected from each block. Ultimately we will have 12 villages from all over India from which we can take respondents for the final sample.

Advantages:

. More representative
. Saves cost
. Complete listing of population is not necessary.

Non-Probability sampling: Mass media researchers frequently use non-probability sampling.
This is the kind of sampling where all sampling units don’t have the equal chance of selection.

Types: The various types of non-probability sampling are given below:

Convenience sampling
Purposive sampling
Quota sampling
Snowball sampling
Volunteer sampling

Convenience/Available sampling: This is known as accidental or haphazard sampling. This is a collection of readily accessible subjects for study.

Example: During election times, media personnel often present man-on-the-street interviews that are presumed to reflect public opinion.

Convenience sampling is best suited for exploratory research which becomes the base for further investigations.

Advantages:

. Quick and economical

Disadvantages:

It may be biased as:
. The respondents may have a vested interests to serve in co-operating with the interviewer.
. Respondents may be those who are vocal or want to brag.

Purposive sampling: It is also known as judgmental sampling. It includes subjects or elements selected for specific characteristics or qualities and eliminates those who fail to meet those criteria.

Example: Often used in advertising studies where researchers select the subjects who use a particular type of product and ask them to compare it with a new product.

Disadvantages:

. Not representative in nature.

Quota sampling: Subjects are selected to a predetermined or known percentage/quota.

Example: A researcher interested to find out how DTH service takers are different from non-DTH service takers in their use of TV may know that 10% of a particular population avail DTH services. The sample the researcher selects, therefore, would be composed of 10% of DTH service takers and 90% of non-DTH service takers.

Advantages:

. Less costly than other techniques.
. Does not require sampling frame.
. Relatively effective and can be completed in a very short time.

Disadvantages:
. It might not be representative.
. It might have interviewer’s bias in selection.
. Strict control of fieldwork is difficult (instead of 25 only 20 respondents may be available.)

Snowball sampling: In this technique, the researcher begins the research with a few respondents who are known and available to him. Subsequently, the respondents give other names who meet the criteria of research, who in turn give more names. This process is continued until adequate numbers of persons are interviewed or until no more respondents are discovered.

Example: A research on the sexual behaviour of homosexuals in a conservative society.

This method is employed when the target population is unknown or when it is difficult to approach the respondents in any other way.

Advantages:

. Reduced sample size
. Reduced cost

Disadvantages:

A person known to someone has a higher probability of being similar to the first person.

Volunteer sampling: This is a technique in which the respondents themselves finally volunteer to give information they hold.

1 comment:

Unknown said...

good
http://business.afu.ac.ae/en/Business-video-gallery/
http://www.afu.ac.ae/ar/admission/afu-general-admission-requirements/
http://www.afu.ac.ae/ar/admission/graduate-admissions/
http://www.afu.ac.ae/ar/tuition-and-fees/