Sampling and

Sampling
techniques

EPI 100 Biostatistics and Epidemiology

MLS 3
By: Enoch Caryl B. Taclan BSBiol MScBiol*
OUTLINE
• Definition of Sampling
• Major Processes and Advantages of Sampling
• Basic Concept Used in Sampling
• Two Types of Population
• Basic Sampling Design
• Types of Non-Probability and Probability Sampling
• Sampling Size determination
 Sampling
- is a process which involves
the study or examination of a
segment of a population to
represent the whole.
How sampling
is applied in
daily life?
How sampling
is applied in
daily life?
TWO MAJOR PROCESSES OF
SAMPLING

1. Getting a segment from the population

- “Segment”
2. Gather or generalize information about
the population
- “Study and examine”
 TWO MAJOR PROCESSES OF
SAMPLING

✓ When doing sampling, you need to remember that your sample must be
representative of your population.

✓ The size of your sample must be adequate enough.

“What is the use of taking only a segment of the population out of the
whole? “

“Is this really valid?”

ADVANTAGES OF SAMPLING

1. It is cheap.
➢ Imagine how much it will cost studying
a whole population?
Example: Nationwide survey on the
effectiveness of hybrid classes among private school
learners? Sample or Population?
ADVANTAGES OF SAMPLING

2. It is faster to conduct.
➢ Observe a “portion”
ADVANTAGES OF SAMPLING

• It provides better quality of information.

• Example: Imagine that you are given a
pile of recommended readings for your
RRL or Introduction in your study.
ADVANTAGES OF SAMPLING

• More comprehensive data can be

gathered.
• Example: Member of National Nutrition
Council you get the population or
sample?
ADVANTAGES OF SAMPLING

• It is the only possible means to gather

information that require destructive
procedures.
Example: Determine the RBC and WBC
count of a patient, you get the population
or sample?
Basic Concepts
Used in Sampling
Population
- Entiregroup of Two Types of
individuals or items of Population
interest in the study.
a. TARGET POPULATION
- Group from which representative
information is desired
− Group from which inferences will be
made
b. SAMPLING POPULATION
- Group from which sample will actually
be taken
SAMPLING POPULATION
❖ Sampling Frame - Collection of
sampling units

❖ Sampling Unit - Units selected from

the sampling population

❖ Elementary Unit- Object or person

from which a measurement is actually
taken or an observation is made
Example:
Objective: To determine the level of
satisfaction of students with the
library services at MAC
Population: Students
Target Population: All Students
studying in MAC
Sampling Population: Students who
utilizes the library
Sampling Frame: A list of students
from different colleges who utilize
library services
Sampling Unit: After that, we can
now identify our sampling unit.
Elementary Unit: Individual student
who will get interviewed or surveyed
for the study.
Basic Sampling
Design
Non-Probability Sampling
- The probability of ❖ Judgment or Purposive
each member of the Sampling
population to be ❖ Accidental or Haphazard
selected/chosen as Sampling
a part of the
sample is difficult ❖ Quota Sampling
to determine or ❖ Snowball Sampling
cannot be specified.
Probability Sampling
- Eachmember of the population ❖ Simple Random Sampling
has a known non-zero
❖ Systematic Sampling
chance of being selected as a
sample. ❖ Stratified Random
Sampling
− Every member has an equal
chance of being chosen ❖ Cluster Sampling
because of the concept of ❖ Special form of sampling:
chance or randomness. Multi-stage Sampling
TYPES OF NON-
PROBABILITY
SAMPLING
DESIGNS
❖ Judgement or
Purposive Sampling
• In this sampling design, a
“representative sample” of the
population is selected based on an
expert’s subjective judgment or on pre-
specified criteria
• Example:
Objective: To determine the knowledge,
attitudes, and practices of traditional healers about
Herbal Medicines
❖ Accidental of
Haphazard Sampling
• In this sampling design, the sample is
usually consisting of those who come at
hand or who is available.
• Example:
Objective: To determine the likelihood of
shoppers to patronize a new market product via ambush
interviews or selecting passers-by to answer
questionnaires.
❖ Quota Sampling
• A quota or pre-determined
subdivision is set, then samples are
identifies until the quota is met.
• Example:
Objective: To determine the brand preference
of ecigarette-users in a particular city, the researchers
set a quota of 100.
❖ Snowball Sampling
• Sample is obtained by a process
whereby an individual to be included
is identified by a member who
waspreviously included.
• Example:
Objective: To determine the prevalence of drug
abuse among young adult aged 25 to 35 in Pasay City,
Manila
TYPES OF
PROBABILITY
SAMPLING DESIGNS
❖ Simple
Random - It is the simplest type of sampling
design.
Sampling − Every element in the population has an
equal chance of being included in the
sample.
❖ Simple
Steps to follow:
i. Construct the sampling frame.

Random
ii. Calculate the sample size.
iii. Select the required number of

Sampling
sampling units using either the
lottery method or random numbers
generated by a calculator or
computer applications.
Example: Population of teachers from
the MLS department of SOAH

• STEP 1: Construct Sampling Frame

- From a population of 11, we will
only draw a sample of 3
• STEP 2: Calculate the sample size
- Sample size of 3
• STEP 3: Select the required
number of sampling units using
either the lottery method - Pick 3
random papers
 - Variation of Simple Random Sampling
− Used when sampling frame is unavailable,
or population is

❖ Systematic
too large to be arranged or numbered
chronologically
− It requires the computation for the

Sampling sampling interval, k

k = N/n
Where:
N = total population size
n = sample size
 Steps to follow:

❖ Systematic
i. Compute for the sampling interval, k.
ii. Draw a number from 1 to K. (This

Sampling
will be the starting point)
iii. Count k intervals from the starting
point until the sample size is
reached.
Example: Population (30) of corporate
workers in a very big industry n=10

• STEP 1: Compute for Sampling

Interval- The first step is to compute
for the sampling interval (k). 30/10 = 3
• STEP 2: Draw a number from 1 to k.
1 to 3 because 3 is our k. And 3 will
be our starting point
• STEP 3: Count k intervals from the
starting point until the sample size is
reached.
Our interval is 3 so we will count 3. We will
count by three until reach the sample size
of 10.
 - The population is first divided into non-
overlapping groups
called stratum

❖ Stratified − This ensures the researcher that the

subgroups considered

Random relevant to the study are adequately

represented

Sampling − It requires the computation for the

sampling fraction, p
p = n/N
Where:
N = total population size
n = sample size
❖ Stratified
Steps to follow:
i. Divide the population into strata.

Random
ii. Obtain a sampling frame from each
stratum.

Sampling
iii. Compute the sampling fraction, p.
iv. Select random samples of p in each
stratum
 Example: The industry you want to determine which
among the girls and the boys will have higher uric acid levels.
We have a population of 50 workers
• STEP 1: Divide the population into strata.
Boys = 24 boys, Girls = 26 girls
• STEP 2: Obtain a sampling frame from
each stratum.
• STEP 3. We compute for the sampling
fraction (p). 10/50 = 0.2.
Now, to get how many people will serve as
our sample, We need to multiply the (p) to
these numbers: 24 and 26.
0.2 x 24 = 4.8 = 5 boys (rounded-off)
0.2 x 26 = 5.2 = 5 girls (rounded-off)
STEP 4. Select random samples of p in each
stratum
This 5 girls and 5 boys are now your
sample.
 - In this design, the sampling units are

❖ Cluster cluster of elements

− Clusters are usually of the same size

Sampling and the characteristics of units across

clusters are similar.
 Steps to follow:

❖ Cluster
i. Divide the population into clusters
ii. Perform Simple Random Sampling

Sampling
iii. Every element in the sampling unit
drawn as sample may or may not be
included in the study.
 Example: Determine the potability of water sources in
Barangay Malinis. And then in Barangay Malinis, there are
four puroks.

• STEP 1: Divide the population into

clusters.
We cluster Baragay Malinis using the
four puroks: Purok 1, Purok 2, Purok 3, and
Purok 44
• STEP 2: Perform Simple Random
Sampling
• STEP 3. Every element in the sampling
unit drawn as sample
So in this case, in cluster sampling, the
sampling unit and elementary unit are different:
Sampling Unit = Purok/Street
Elementary Unit = Household
❖ Multi- - Very special form of sampling design
− A procedure carried out in phases

stage − Usually involves more than one

sampling method

Sampling − Often used on wide coverage studies.

Usually in national
surveys
❖ Multi-
Steps to follow:
i. Divide the population into sets of

stage
primary sampling units.
ii. Selected primary sampling units are

Sampling
divided into secondary sampling
units.
iii. The procedure continues until the
desired stage is reached.
 Example: National Nutrition Survey: The country has 13
regions.

• STEP 1: Divide the population into sets of

primary sampling units.
Choose the 13 regions
• STEP 2: Selected primary sampling units
are divided into secondary sampling
units.
From the 13 regions, you will perform
sampling to choose two provinces.
• STEP 3. The procedure continues until
the desired stage is reached

Measure the weight and height of 0 to

19 year old healthy individuals to
identify the nutritional status.
SAMPLE SIZE
DETERMINATION
SLOVIN’S FORMULA
To determine the sample size, we
well use this formula.
Where:
n = Sample Size
N = Population Size
e = Margin of Error
Example
A group of researchers will conduct a n =20,000 / 1 + (20,000)(. 10)^2
survey to find out the opinion of mothers n = 20,000 / 1 + (20,000)(. 01)
in Municipality Masagana regarding the
price hike on rice products. If there are n = 20,000 / 201
20,000 residents in the community and = 99.502 or 99.50 or 100
the researchers plan to use a sample
using a 10% margin of error. What should
the sample size be?
WRAP UP

➢ The process of sampling involves getting

representatives from the population as a sample.
➢ Then, studying the sample in order to generalize
information or make inferences or applications to the
total population.
WRAP UP
➢ Population - the entirety, which can be a target population
or the group of individuals from which inferences will be
made after doing your study.
➢ Sampling Population - group from which a sample will be
taken.
➢ Sampling Units - units selected from the sampling
population which you can usually get through the sampling
frame.
➢ Sampling Frame - collection of your sampling units.
➢ Elementary Unit - an object or a person from which a
measurement is taken or an observation is made.
WRAP UP
Non-Probability Probability Sampling
Sampling ➢ Everyone in the population has a known
➢ It is difficult to non-zero chance to be included in the
determine the chance sample.
of every member of the ➢ There is the concept of chance and
population to be randomness.
included as a sample. ➢ Simple Random Sampling
➢ Judgment or ➢ Systematic Sampling
Purposive Sampling
➢ Stratified Random Sampling
➢ Accidental or
Haphazard Sampling ➢ Cluster Sampling
➢ Quota Sampling ➢ Special form of sampling: Multi-stage
Sampling
➢ Snowball Sampling
 WHICH SAMPLING METHOD IS BEST?
➢ It will always depend on the objective/s of the research/
researchers.
➢ There is no perfect sampling design as long as it will help
the researchers to obtain or fulfill the research objective.
All sampling designs can be used.
 CRITERIA OF A GOOD SAMPLING
DESIGN
➢ With a sample that is “representative of the population”
➢ With adequate sample size.
➢ Practical and feasible
➢ Economic and efficient
Thank you!