Professional Documents
Culture Documents
METHODS/TECHNIQUES
2.1 Quota sampling
1. Simple random 2.2 Snowball sampling
2. Stratified random 2.3 Purposive, judgmental or
3. Cluster sampling deliberate sampling , Genealogy
4. Systematic sampling Based Sampling
5. Multi stage sampling 2.4 Convenience sampling
2.5 Matched Sampling
Probability Sampling
- Probability sampling means that every item in the
population has an equal chance of being included in
sample. One way to undertake random sampling would
be if researcher was to construct a sampling frame first
and then used a random number generation computer
program to pick a sample from the sampling frame
(Zikmund, 2002).
- Probability or random sampling has the greatest
freedom from bias but may represent the most costly
sample in terms of time and energy for a given level of
sampling error (Brown, 1947).
Probability Sampling
- Advantages:
o This sampling technique reduces the chance of
systematic errors.
o The methods minimize the chance of sampling
biases.
o A better representative sample is produced using
probability sampling techniques.
o Inferences drawn from sample are generalizable to
the population.
- Disadvantages:
o The techniques need a lot of efforts
o A lot of time is consumed.
o They are expensive.
Probability Sampling
1.1. Simple random sampling
- The simple random sample means that every
case of the population has an equal
probability of inclusion in sample.
Disadvantages associated with simple
random sampling include (Ghauri and
Gronhaug, 2005):
Probability Sampling
1.1. Simple random sampling
a. Method:
i. Before taking a sample the population is needed to be defined. In
other words, one must know what characteristics constitute the
population of interest.
ii.A list of all the elements of population is required.
iii.One needs to prepare the list if the readymade is not available.
iv.The list must be exhaustive i.e. it must contain the name of each and
every element of the population.
v.One method for the selection of participants is lottery method: each
element is first given a number and then numbers are individually
written on slips of paper. The slips are put and mixed thoroughly in
some bag or bowl. Then the decided number of slips is drawn out of it.
vi.Other methods may be the use of any random table generated
through computer or any other resource.
vii.The selected participants are approached and investigation is done.
Probability Sampling
1.1. Simple random sampling
b. Benefits:
i. There is not the possibility of sampling biases.
ii. The sample is a good representative of the population.
c. Crucial Issues/ Draw Backs:
iii.It may be very costly and time consuming especially in those cases
when the participants are widely spread geographically and difficult
to approach
iv.It needs a lot of efforts especially for a large population.
v. In many circumstances it is not possible to get or prepare an
exhaustive list of elements. Even apparently complete lists may
also exclude some of the potential elements. For instance, we
randomly select the population of a town using telephone list. Is it
sure that everybody in the town has a connection?
Probability Sampling
1.2. Stratified random
sampling
- Stratified sampling is where the population is
divided into strata (or subgroups) and a
random sample is taken from each subgroup. A
subgroup is a natural set of items. Subgroups
might be based on company size, gender or
occupation (to name but a few). Stratified
sampling is often used where there is a great
deal of variation within a population. Its
purpose is to ensure that every stratum is
adequately represented (Ackoff, 1953).
Probability Sampling
1.2. Stratified random
sampling
- The sample is drawn in such a way as to
ensure adequate representations for each of
the population sub-groups or strata in the
sample. After sample is determined, sample
size for each group is computed using the
formula:
Where:
nt n1 = sample size for first group
N1 = population of first group
Nt = total population
nt = total sample size
Probability Sampling
1.2. Stratified random sampling
a. Method:
i. Before taking a sample the population is needed to be defined. In
other words, one must know what characteristics constitute the
population of interest.
ii. On the basis of nature and purpose of investigation it is decided
which criterion has to be taken into account to make the sub groups
(strata)
iii. Then on the basis of decided criterion stratification is done
iv. A list of all the elements of each strata is required.
v. One needs to prepare the list if the readymade is not available.
vi. The list must be exhaustive.
vii. The participants are then selected from each stratum through
lottery method or using any random table (as in simple random
sampling)
Probability Sampling
1.2. Stratified random sampling
b. Benefits:
i. For a heterogeneous population it produces a representative
sample as it captures the diversity which otherwise is likely to be
undermined through simple random or systematic random
sampling.
c. Crucial Issues/ Draw Backs
i. It needs a lot of efforts.
ii. It is costly and time consuming
iii. If the criterion characteristic/ variable used for classification is
not selected correctly, the whole research may go in vain
Probability Sampling
1.3. Cluster sampling
- Cluster refers to any entire group of
individuals or objects which share similar
characteristics important in the study. The
samples are drawn in clusters rather than in
using separate individuals. It is the groups not
the individuals that are randomly selected.
Grouping of individuals is based on the
different categories/ characteristics like, the
geographical setting where individuals belong,
commodity raised, tribal affiliation, etc.
Probability Sampling
1.3. Cluster sampling
- Cluster sampling is where the whole
population is divided into clusters or groups.
Subsequently, a random sample is taken from
these clusters, all of which are used in the final
sample (Wilson, 2010). Cluster sampling is
advantageous for those researchers whose
subjects are fragmented over large
geographical areas as it saves time and money
(Davis, 2005).
Probability Sampling
1.3. Cluster sampling
a. Method
i. First of all the population is divided into clusters.
ii. The clusters are selected randomly using simple random or
systematic random sampling techniques.
iii. The selected clusters are visited.
iv. All the elements (may be individuals, households, schools,
markets etc. depending on the nature of investigation) within
the selected clusters are investigated
Probability Sampling
1.3. Cluster sampling
b. Advantages
i. More convenient for geographically dispersed populations
ii. Reduced travel costs to contact sample elements
iii. Simplified administration of the survey
iv. Unavailability of sampling frame prohibits using other random sampling
methods
v. In cases where the population is spread over a wide geographical region,
cluster sampling is used to reduce cost as compare to simple random or
systematic random sampling.
vi. It consumes less time and efforts than the aforementioned techniques.
vii. For instance the list of elements of the population is not required.
viii. Moreover, instead of going place to place over a widely spread area for
randomly selecting elements, you get a group of elements in one
geographical region.
Probability Sampling
1.3. Cluster sampling
c. Disadvantages
i. Statistically less efficient when the cluster elements are similar
ii. Costs and problems of statistical analysis are greater than for
simple random sampling
d. Crucial Issues/ Draw Backs:
i. It may sometimes lead to sampling biases and systematic errors.
ii. For instance, in the selection of markets only big markets may
be selected, though randomly. So this selection is likely to impact the
results. The results may be different if there were small markets in
the selection too.
iii. If clusters are not homogeneous among them, the final sample may
not be representative of the population.
Probability Sampling
1.4. Systematic sampling
- Systematic sampling is where every nth case
after a random start is selected. For example,
if surveying a sample of consumers, every fifth
consumer may be selected from your sample.
The advantage of this sampling technique is its
simplicity. The sampling interval is computed
before selecting the right samples with the
formula:
Where: K = Size of selection interval
N = Population Size
n = Sample Size
Probability Sampling
1.4.Systematic sampling
a. Method:
i. Before taking a sample the population is needed to be defined. In
other words, one must know what characteristics constitute the
population of interest.
ii. In case where exhaustive list of elements of the target population is
available, the list is arranged and numbered in an order 1 to N.
iii. To find an appropriate interval suppose population contains N
number of elements and we need a sample of n size.
iv. Divide N by n. the number obtained through this division, say k, is
an appropriate interval size to produce a representative sample.
v. Ask for a starting number from a person(1 to Nth)
v. In case where the list is not possible to make an interval size is
decided and then participants appearing with that interval are
approached.
Probability Sampling
1.4.Systematic sampling
b. Benefits:
i. It ensures the extension of sample to the whole population
ii. It provides the way to get a random and representative
sample in the situation where prior listing up of elements is
not possible.
c. Crucial Issues/ Draw Backs:
i. It may be very costly and time consuming especially in
those cases when the participants are widely spread
geographically and difficult to approach
ii. It needs a lot of efforts especially for a large population.
Probability Sampling
1.5. Multi-stage sampling
- This sampling involves several phases in
drawing the samples from the population. The
population units are grouped and arranged
into hierarchical order or level, and sampling is
done successively.
- Multi-stage sampling is a process of moving
from a broad to a narrow sample, using a step
by step process (Ackoff, 1953).
Probability Sampling
1.5. Multi-stage sampling
a.Method:
i. Usually at the first stage target population is divided into clusters.
ii. The clusters are selected randomly.
iii.These clusters are called as first stage units or primary units
iv. These clusters are homogenous among them but may be
heterogeneous inside.
v. To overcome this heterogeneity, homogenous sub groups called as
strata are formed.
vi.So the strata are called the second stage units or sub-units.
vii. The formation of these strata can be done using cluster
sampling technique or stratified random sampling technique
depending on the nature of investigation.
viii. In each stratum the units may need to be further divided, for
instance market places into shops, buildings into houses etc.
ix.The final units obtained are investigated.
Probability Sampling
1.5. Multi-stage sampling
b. Advantages:
i. It increases cost and time efficacy.
ii. For instance instead of investigating all the elements within a cluster, if a
sample is randomly drawn from each cluster, the results will be similar but
with lesser efforts. This particular sampling method where random sampling
is done within the selected clusters is called as two stage sampling.
iii. The technique is also useful in overcoming the heterogeneity problem
within the clusters.
c. Crucial Issues/ Draw Backs:
i. If the selected clusters do not capture the characteristic diversity of
population, the sample would not be representative of the population.
ii. If the characteristic variable used for making strata (in case of
heterogeneity) at any stage is not appropriately selected depending on the
nature of investigation, the whole research may go in vain.
Non Probability Sampling
- Non probability sampling is often associated
with case study research design and
qualitative research. With regards to the
latter, case studies tend to focus on small
samples and are intended to examine a real
life phenomenon, not to make statistical
inferences in relation to the wider population
(Yin, 2003).
- A sample of participants or cases does not
need to be representative, or random, but a
clear rationale is needed for the inclusion of
some cases or individuals rather than others.
Non Probability Sampling
Advantages:
• The techniques need less effort.
• These techniques need less time to finish
up.
• They are not much costly.
Disadvantages:
• The sampling techniques are prone to
encounter with systematic errors and
sampling biases.
• The sample cannot be claimed to be a
good representative of the population.
• Inferences drawn from sample are not
generalizable to the population.
Non Probability Sampling
2.1. Quota sampling
- The samples to be gathered depend on the
needed participants. It is assumed that the
samples will match the population with regard to
the chosen set of characteristics (Vockell, 1983).
- Quota sampling is a non-random sampling
technique in which participants are chosen on the
basis of predetermined characteristics so that the
total sample will have the same distribution of
characteristics as the wider population (Davis,
2005).
Non Probability Sampling
2.1.Quota sampling
a. Method:
i. First of all there is a need to identify the variable which makes the
target
ii. population heterogeneous.
iii. On the basis of the identified variable sub groups are made.
iv. A quota is set for each sub group.
v. Then the sample is approached on the basis of set quota
b. Advantage:
i. It ensures the presence of every sub group of the population in the
sample.
ii. There is not the requirement of any lists of the elements of population.
iii. It is less time consuming and low in cost than stratified random
sampling.
c. Crucial Issues/Draw Backs:
Non Probability Sampling
2.1.Quota sampling
a. Method:
i. The investigator selects a person who matches the criteria of the
research
ii. The first participant is now asked to refer the investigator to
another person who meets the same criteria.
iii.Now the second participant approached is asked to refer the
researcher to another one. In this way a chain is made.
b. Advantage:
i. This technique is useful in approaching the type of population
which is not readily available or present in a very small quantity.
c. Crucial Issues/Draw Backs:
i. It is subjected to sampling biases and systematic errors due to
network connection.
Non Probability Sampling
2. 3. Purposive, judgmental or deliberate
sampling