SAMPLING

ADRIAN P. YBANEZ, DVM, CBO, RMicro, DRDM, MBA, MCDRM, DipPCVPH, PhD
 Sampling
The act of studying or examining only a segment of the
population to represent the whole. Whatever
findings we get for this segment of the population,
we generalize to the total population.
e.g., taste a grape before we buy whole bunch,
prevalence of drug and substance abuse in high
school students – study only certain students,
prevalence of diabetes among Cebuanos – study only
certain Cebuanos.
 Advantages of Sampling
1. Cheaper
2. Faster
3. Better quality of information. A smaller
group being studied requires only a smaller
number of data collectors who can be
trained more rigidly and supervised closely.
Allows the use of more complicated and
accurate technical methods. E.g., prevalence
of anemia – blood samples of determining
hemoglobin vs. pallor of the conjunctiva.
4. More comprehensive data can be obtained.
 Uses of sampling
1. Evaluate the health status of a population. E.g.,
prevalence surveys
2. Investigate the factors affecting a dependent
variable (health; risk factors for given diseases
or determinants of certain practices like
breastfeeding).
3. Evaluate effectiveness of health measures. To
stop, modify or continue existing public health
programs
 Definition of Terms
1. Population – entire group of individuals or
items of interest in the study. People,
medical records, houses.
2. Target population – group from which
representative information is desired and to
which inferences will be made. Whatever
conclusions that will be derived from the
study will be generalized to the target
population.
 Definition of Terms
3. Sample
Subset of representative units from the
population.
Used to represent the population for statistical
study
Findings from the sample are used as the basis for
estimating or predicting the characteristics of
the population.
4. Sampling Frame:
Actual list of each element or member of the
target population from which the sample may be
drawn.
 Definition of Terms
5.Element or elementary unit – an object or a person
on which a measurement is actually taken or an
observation is made. If, in an environmental
sanitation survey, the presence of a sanitary toilet
and a potable water supply of each sample
households are noted, then the household is the
element.
6. Sampling unit. Refers to the units which are chosen
in selecting the sample, and may be made up of
non-overlapping collection of elements or
elementary units.
 List of Elements
Sampling Frame 1
2
3
4
3 4 5 5
2 6 6
1 11 10 7 7
12 9 8 8
9
13 10
14 16 17
11
15 12
13
14
Sample 15
16
17
The Sampling Process

Define the Target Population

Identify the Sampling Frame

Choose the Sampling Method

Determine the Sample Size

Gather the Data

 Choosing the Sampling Method
• Probability Sampling: Every element in the
population has a known, nonzero probability of
being selected. The rules and procedures for
selecting the sample are explicitly and rigidly
specified.

• Nonprobability Sampling: Probability of any

particular member of the population being chosen
in the sample is difficult to determine or cannot be
specified.
 Types of Sampling Methods
Sampling Methods

Probability Sampling Non-probability Sampling

Accidental
Simple Strati- Judge-
Systematic Cluster Or Quota Snowball
Random Fied Ment
Sampling Sampling Haphazard Sampling Sampling
Sampling Sampling Sampling
Sampling
 Non-probability sampling
• Judgment or purposing sampling.
• Accidental or haphazard sampling
• Quota sampling
• Snowball sampling
Judgment or purposing sampling
“A representative” sample of the population is
selected based on an expert’s subjective
judgment or on some pre-specified criteria.
For example, study areas are frequently
selected in a purposive manner based on their
proximity to the office conducting the
research, or the level of cooperation of the
people living in the area.
Accidental or haphazard sampling.
A researcher uses in his study whatever items
come at hand or whoever is available. He may
interview the first 50 people who enter a
department store or he may ask for
volunteers.
 Quota sampling
• Data collectors are given quotas to meet.
Interviewers may be given instructions to keep
on interviewing household heads in a given
place until the pre-specified data for that
place has been reached.
 Snowball sampling
Devised because of the difficulty of producing a sampling
frame and identifying members of the population. The
first person identified to be a member of the target
population will be interviewed for the study and will be
asked to identify other members of the population.
Those identified by him will be the next respondents and
will likewise be asked to identify other members. The
process continues until the desired sample size has been
reached. Example, a survey of use of contraceptives
among commercial sex workers.
 Probability sampling
1. Simple random sampling.
2. Systematic sampling.
3. Stratified random sampling.
4. Cluster sampling
5. Multi-stage sampling design.
 Simple Random Sampling
Most basic type of sampling design.
Every element in the population has an equal
chance of being included in the sample.
Eg. Rolling identical pieces of paper with the
name of each population member in it, and
then drawing some of the names after the
pieces of paper are thoroughly mixed in a box
or tambiolo.
When the population is large, in order to select a
sample by SRS, a chronologically numbered
listing of the population is required to serve as
the sampling frame.
This is difficult to do for surveys of wide coverage
such as nationwide survey.
SRS is therefore used only in surveys involving
relatively small populations with readily
available sampling frames.
 Systematic sampling
Sampling interval
k= population size (N)
sample size (n)
Example, to get a sample of 10 households from a
population of 80, k = 80/10 = 8.
After k has been determined, one number from 1 to k will
be drawn at random. The population element
corresponding to that number and every kth number
thereafter will be included in the sample.
Example: population of 80 households.
sample of 10 houses
k=8
Draw any number between 1 and 8. This means
that household number 5 in the frame will be
the first to be included in the sample.
All others include every 8th house thereafter, i.e.,
those households with the following numbers
in the frame: 13, 21, 29, 37, 45, 53, 61, 69, and
77.
Systematic sampling may be resorted to when a
frame is not available. For example, households
in a newly developed subdivision or village may
not appear in existing lists or maps. Instead of
going through the process of constructing a
frame, it may be more convenient and practical
to instruct the interviewers to select at random
one house from the first k houses, say 10 houses
that they find in the village, and interview in
every 10th house thereafter.
Systematic sampling
-Used when the sampling units are too numerous to
number for purposes of SRS.
-This may be the case when one selects a sample of
medical charts or a sample of birth and death
certificates. Since there are thousands of these,
numbering each one will be too time consuming
and tedious. One can just count verbally as he
goes along, or use a ruler (the number of records
per inch or any unit length is determined) and
then apply the principles of systematic sampling.
 Stratified random sampling
The population is first divided into non-
overlapping groups called strata.

A simple random sample is then selected from

each stratum. A common reason for this type
of sampling is to ensure that the subgroups
are adequately represented
• Stratification will always achieve greater
precision provided that the strata have been
chosen so that members of the same stratum
are as similar as possible in respect of the
characteristic of interest.
• Strata:
1. social classes
2. educational attainment
3. age groups
 Cluster sampling
When a sampling frame for the elements is not
readily available or when cost considerations
are important.
The sampling units are clusters of elements.
For example, a census block may be considered
a cluster of households, a hospital is a cluster
of employees, a class is a cluster of students.
• This type of sampling involves the
following steps:
–The population is first divided into
clusters that serve as the sampling
units and a sample of units is selected.
–Every element found in each sampling
unit drawn as sample may or may not
be included in the study.
DIFFERENCE BETWEEN CLUSTER SAMPLING
AND STRATIFIED SAMPLING

In stratified random sampling, all the strata of

the population is sampled while in cluster
sampling, the researcher only randomly
selects a number of clusters from the
collection of clusters of the entire population.
Therefore, only a number of clusters are
sampled, all the other clusters are left
unrepresented.
An example of cluster sampling is selecting a
random sample of classes in a school survey. All
students in each class selected in the sample will
be included in the study. In this design, each
class represents a cluster or group of students,
hence, the term “cluster sampling.” Another
characteristic of cluster sampling is the difference
between the sampling unit and the elementary
unit. In this example, the sampling units are the
classes while the elementary units are the
students.
 Multi-stage sampling design
• Used when the sample survey to be conducted has a wide
coverage such as in nationwide survey. The following steps
are done:
– The population is first divided into a set of primary or
first-stage sampling units. A sample of such units is
selected.
– Each primary sampling unit included in the sample is
further subdivided into secondary or second-stage
sampling units from which a sample will again be taken.
– The procedure continues until the desired stage is
reached.