Module 3+4
11 May 2023
20:02
SAMPLING
o The basic idea of sampling is that by selecting some of the elements in a population,
we may draw conclusions about the entire population.
o A census involves collection from every member of the population being studies
o A population is the aggregate of all the members of a defined group that is being
studied
o For example, each staff member questioned about an optimal promotional strategy
is a population element, each advertising account analyzed is an element of an
account population, and each ad is an element of a population of advertisements
STAGES OF SAMPLING
1. What is the target population?
o Target population is a complete or theoretical segregation of study elements. It is
the population to which research results are ideally to be generalized
o The target population is typically defined in terms of element, units, time, and
geography. Elements is the object about which the information is collected and that
provides the basis of analysis. Units are set of elements considered for selection in
sampling
2. What are the parameters of interest?
o Population parameters are summary descriptors (e.g., incidence proportion, mean,
variance) of variables of interest in the population. Sample statistics are descriptors
of those same relevant variables computed from sample data
3. What is the sampling frame?
o Sampling frame is the actual list of sampling units or elements from which the
sample is selected.
o When a list is referred to as a sampling frame, it is possible that the list does not
have all the elements of the target population.
o It is also possible that the list has elements that do not belong to target population.
Such incongruity can lead to an error called sampling frame error.
4. What is the appropriate sampling method?
o The next step is the method of selection of sample. The choice of sampling methods
is usually dictated by feasibility in terms of time and resources.
5. What size sample is needed?
o Sample size is the number of elements that will be included in the sample for study.
TYPES OF SAMPLING
1. PROBABILITY
o In probability sampling, sample is selected from a population based on the principal
of randomization. The elements from the population are randomly selected. Each
element's probability of inclusion can be calculated
1. Simple random sampling
Simple random sampling is defined by two properties.
i. Each member of the population has an equal and known probability of being
selected in the sample
ii. Each combination of members of the population has an equal chance of forming the
sample
The technique requires a complete list of the people of the target population and their
contact details. No other information with regards to sampling frame, geography, etc is
required
The disadvantage of simple random sampling is that even if the researcehr knows that the
population is made up of 50 indian cusines and 50 international cuisines, no use is made of
the information.
2. Systematic sampling
Also called interval sampling
The sample is chosen by selecting a random starting point and then picking every ith
element in succession from the sampling frame
The intervals under is rounded to the closest integer
If the ordering of the elements is related to the characteristics of the interest, systematic
sampling enhances the representativeness of the sample
If the ordering of the elements if in cyclical pattern, systematic sampling may reduce the
representativeness of the sample
3. Stratified sampling
For the stratified sampling, the population is divided into one homogenous groups called
strata. The strata should be mutually exclusive and collectively exhaustive.
Independent samples are selected from each stratum using any sampling method
The sampling method can be different for different strata
If the simple sampling method is used to select the sample from each stratum then it is
called stratified simple sampling method
Needs larger sample to get more accurate information
It ensures that an adequate sample size from sub-groups in the population is taken
4. Cluster sampling
In this, the target population is first divided into subpopulation called clusters. The clusters
are mutually exclusive and collectively exhaustive subpopulations.
A random sample of clusters is selected, using probability sampling techniques such as
systematic sampling or simple random sampling
When each cluster being included in the sample is selected randomly such that the
probability of selection is proportional to the size of the cluster, it is called sampling with
probability proportional to size
If the clusters are geographic sub-divisions, then the cluster sampling is also referred to as
area sampling
1. NON-PROBABILTY
o Under the non- probability sampling the researcher purposively chooses the
particular elements of the population for constituting a sample
1. Convenience sampling
Convenience sampling is done based on the ease of selection of sample. The samples are
chosen simply because they were the most convenient to choose
The results from such samples cannot be generalised to the population
Example: shoppers at a single mall
Sample may not represent the whole population
2. Purposive sampling/judgmental sampling
For purposive sampling, the researcher chooses the sample based on who they think would
be appropriate for the study
Also called judgmental sampling
In such kind of sampling, great chances of personal bias.
The researcher may select sample which will yeild favorable results to their view point
3. Quota sampling
Under the quota sampling the samples are selected in quotas from different strata
The actual selection of the items for the sample Is based on the reseacher's discretion.
It may also be considered two-stage restricted judgmental sampling.
Sample is choosen to include specific proportions of elements
Selection of sample is totally based on researcher's judgment as long as the quota criterias
are met
4. Snowball sampling
Snowball sampling is the non-probability sampling technique where existing sample
elements recruit future sample elements from among their acquaintances
Also called chain sampling, chain-referral sampling, or referral sampling
The sampling starts with small pool of elements that fit the sampling criteria, later their
referral the sample size grows
Examples can be alcohol addicts or AIDS patients
SAMPLING ERRORS
Read from ppt
