SAMPLING AND SAMPLING TECHNIQUES
3/27/2024 1
• A sample is a finite part of a statistical population whose
properties are studied to gain information about the
whole(Webster, 1985).
• A population is a group of individuals such as persons,
objects, or items from which samples are taken for
measurement.
For example a population of presidents or professors, books
or students.
3/27/2024 2
Sampling:
Is a process or technique of selecting suitable sample, or
representative part of the population from the large
population.
Necessary?
Sampling is not always necessary. When a
population is small, we may choose to survey all
its members.
Sampling depend upon population size, what we
want to know, and the resources available.
3/27/2024 3
Why sampling?
Get information about large populations in
Less costs
Less field time
More accuracy i.e. Can Do A Better Job of Data
Collection
When it’s impossible to study the whole
population
3/27/2024 4
Two standard ways to draw a sample.
Probability Sampling Non Probability Sampling
To provide a statistical basis The aim of the evaluation is to
representative sample learn about individuals or
cases for some purpose other
than generalizing to the
population
3/27/2024 5
Sampling for generalizability: probability sampling
• To generalize from the sample to the population, pobability/
random sampling is necessary.
• It provides for random selection of the units.
Implies each unit has equal chance of being selected,
So that it is possible to confidently make estimates about
the total population based on the total sample results, and It
increases the likelihood that the information collected is
representative of the entire group
3/27/2024 6
Sampling
1.Probability sampling 2.Non probability sampling
1.Simple rand sampling 1.Quata sampling
2.Systematic rand sampling 2.Purposive sampling
3.Stratified rand sampling 3. Convenience sampling
4.Cluster rand sampling
3/27/2024 7
Simple random sample
Each unit of the population has an equal chance of
being selected.
This requires a complete list of the total population—
all participants in the program, all homeowners, all
county residents, and so on.
3/27/2024 8
Simple Random Sampling
EX:To evaluate the prevalence of tooth decay among
the 1200 children attending a school.
List of children attending the school
children numerated from 1 to 1200
sample size= 100 children
Random sampling of 100 numbers between 1 and
1200
How to select randomly?
One of the following methods may be selected for
drawing a random sample.
3/27/2024 9
1.Lottery Method:
When the population is small, numbers or
names can simply be drawn from a hat.
Record the number or name that is drawn.
Put the slip back into the container and continue
drawing until the required sample size is obtained.
If the same number is drawn again, disregard it, put
it back in the container and continue.
3/27/2024 10
2 For larger populations :
Random number table may be used (found in most
statistics textbooks)
After assigning consecutive numbers to the names
on the population list, a number is randomly
selected on the table (close eyes and point).
Proceed either vertically or horizontally.
3/27/2024 11
For example, if the total population is a three-digit number
(100), use the last three digits of the random table number
that correspond to a number on the population list, (any
number between 1 and 100) until the needed sample size is
obtained. The corresponding names on the population list
form the sample.
59391 58030 52098 82718 87024
99567 76364 77204 04615 27062
10363 97518 51400 25670 98342
86859 19558 64432 16706 99612
3/27/2024 11258 24591 36863 55368 31721 12
Stratified Sampling
• When population is large, not homogenous,
ie., field survey.
• Population under study is first divided into homogenous
groups according to some common characteristics
called stratas & the sample is drawn from each strata in
proportion to its size.
Ex.: defined areas, classes, ages, sexes etc.
Workers as : Skilled workers & unskilled workers
Religion wise : Hindu, Muslim etc.
3/27/2024 13
SYSTEMATIC SAMPLING
• Used when complete list of population is available.
• To field studies
• Popln is large,scattered / not homogenous.
PROCEDURE:
N=1200, and n=60
sampling interval =1200 = 20
60
List persons from 1 to 1200
3/27/2024 14
Systematic Sampling
• Randomly select a number between 1 and 20 (ex:8)
• 1st person selected = the 8th on the list
• 2nd person=8+20 = the 28th
• etc------
3/27/2024 15
Cluster Sampling
Especially useful for door to door personal surveys
(significantly reduces cost)
When units of population are natural groups or classes as
villages, wards, blocks, slums, children of school etc.
Gives higher standard error. But simpler, involves less time
& cost.
3/27/2024 16
DRAWING THE CLUSTERS
We need:
-Map of the region
-Distribution of the population (by villages or areas)
-Age distribution (population 12-23m 3%)
VILLAGES POP 12-23 cumulative popln
A 53000 1600 1600
B 7300 220 1600+220=1820
C 106000 3200 1820+3200=5420
D “
E “
F “
Total population = 9820
3/27/2024 17
Cluster Sampling
Compute sampling fraction:
k= 9820 = 327
30
Draw a random number between 1 and 327
Ex:62
start from the village including 62 & draw the clusters adding
the sampling fraction.
3/27/2024 18
Drawing households and children
On the Spot
Go to the center of the village, choose direction (randomly)
Number the houses in this direction
Ex:21
Draw random number (between 1 and 21) to identify the first house
to visit
From this house progress until finding the 7 children
3/27/2024 19
Sampling for other purposes: Non probability Samples
In some instances, probability sampling may be
impossible, unnecessary or even undesirable.
When we are limited to some particular
participants, or when we want in-dept
information regarding a particular programme.
The preliminary stages of research or where there
are cost constraints for conducting research.
3/27/2024 20
• In non probability sampling: no expectation that each
unit has an equal chance of being included in the sample.
Since the sample does not intend to represent the
population, findings should not be generalized to the
whole.
Methods:
• Quata sampling
• Purposive sampling
• Convenience sampling
3/27/2024 21
Quata sampling:
A quota sample is that the researcher deliberately
sets the proportions of levels or strata within the
sample.
A quota sample divides the population being
studied into subgroups such as male and female,
younger and older. We estimate the proportion of
people in each group based on what we know
about the population in general.
3/27/2024 22
Example:
If we know that 20% of the population is headed by a
single-parent family, then search for respondents
until 20% of our sample is single parents. This is not
the same as the stratified random sample because
not every single parent is identified and has an equal
opportunity of being included
3/27/2024 23
Convenience sampling ( Availability sampling) :
• It’s carried out on the basis of how easy is it for a
researcher to get in touch with the subjects.
• Is a method of choosing subjects who are available
or easy to find.
• is used when there are time and cost limitations in
collecting feedback.
• In situations where there are resource limitations
such as the initial stages of research, convenience
sampling is used.
3/27/2024 24
.
EX:
• In clinical practice, we might use patients who are available to
us as our sample.
• In many research contexts, we sample simply by asking for
volunteers.
• The primary advantage of the method is that it is
very easy to carry out, relative to other methods
• The problem with this type of sample is no evidence that
they are representative of the populations.
3/27/2024 25
Purposive /Judgment sampling :
Elements are chosen based on purpose of the study.
Example, students who live in dorms on campus, or
experts on urban development.
Serve purposes other than representativeness or
randomness.
Basic to all these is the importance of selecting
cases from which we can learn much about issues that are
important to the study.
Focus on the specific rather than the general.
3/27/2024 26
Example:
If an evaluation’s purpose is to increase a program’s
effectiveness in reaching low-income families, we may
learn more by conducting an in-depth query of the
few poor families in the program than by gathering
sample of all participants.
3/27/2024 27
Summary
• Sampling is the selection of a smaller number
of units from a larger group or population.
• To generalize from the sample to the population,
probability or random sampling is needed to ensure
that the sample is representative.
• If generalizations are not desired or necessary, non
probability sampling is appropriate and can yield very
useful information if cases are selected thoughtfully.
•
3/27/2024 28
Strengths and Weaknesses of
Basic Sampling Techniques
Technique Strengths Weaknesses
Nonprobability Sampling Least expensive,least Selection bias, sample not
Convenience sampling time-consuming, representative, not recommended
most convenient descriptive or causal research
for
JPurposive sampling Low cost, convenient, Does not allow generalization,
not time-consuming subjective
Quota sampling Sample can be
Controlled for certain Selection bias, no assurance of
characteristics representativeness
Probability sampling Easily understood, Difficult to construct sampling
Simple random sampling results projectable frame, expensive, Lower precision,
(SRS) no assurance of representativeness.
Systematic sampling Can increase
representativeness,
easier to implement than
SRS, sampling frame not
necessary
Stratified sampling Include all important Difficult to select relevant
subpopulations, stratification variables, not feasible to
Precision stratify on many variables, expensive
Cluster sampling Easy to implement, Imprecise, difficult to compute and
interpret results
3/27/2024 29
Cost effective
The Sampling Design Process
Define the Population
Determine the Sampling Frame
Select Sampling Technique(s)
Determine the Sample Size
Execute the Sampling Process
3/27/2024 30
