MINDRIANY SYAFILA

TL5003 METODA PENELITIAN

MASTER PROGRAM OF ENVIRONMENTAL ENGINEERING SAMPLING TECHNIQUES
INSTITUT TEKNOLOGI BANDUNG
 INTRODUCTION
A Sample: Sampling Sampling Sampling
population The frame is Element : Unit: is
can be representa the list Each from either a
defined as tive from the single
including all proportion which the population element or
people or of the potential about a collection
items with population responde which of elements
the is called a nts are informatio subject to
characteristi sample drawn n is data
c one wishes collected analysis in
to the sample
understand.

population
Population and Sample sample
 NEEDS OF
PURPOSES OF SAMPLING SAMPLING

Population in many cases may be so large and scattered

that a complete coverage may be not possible.

In a short period of time valid and comparable results ADVANTAGES OF

can be obtained.results can be obtained SAMPLING
 Less resources (time, money)
Gives results with known accuracy that can be  Less workload
calculated mathematically

Sampling is economical since it contains

fewer units
 SAMPLING PROCESS
Principles of Sampling
The main principle of sampling is that the researcher Defining the
population
seeks knowledge about the population by a fewer units
and extend the inference about the sample to the Specifying a
entire population. Steps sampling frame
Specifying a
The principles of sampling are : Involved
sampling method
in the
process of Determining the
sample size
Sample units Sample units Sample units
Sampling Implementing the
must be chosen must be must be sampling plan
in a systematic clearly independent
and objective defined and of each other Sampling and
manner. easily data collection
identifiable.
 SAMPLING TECHNIQUE
PROBABILITY SAMPLING NON PROBABILITY SAMPLING
Information about the Information about the
population available population not available
Random Non Random

Generalizable Not Generalizable

Expensive Less Expensive

Time-consuming More convenient

Less prone to bias Prone to bias

 NONPROBABILITY
SAMPLING TECHNIQUE Merits
• Economical
• Any sampling method where some elements of the • Convenient
population have no chance of selection (these are • Used when time or
sometimes referred red to as ‘out of coverage ‘ / ‘ other factors rather
undercovered ‘), or where the probability of selection than generalizability
can’t be accurately determined. become critical
• The elements do not have a pre-determined chance of
being selected. In this method, samples are not picked
randomly.
• This is usually used when sampling frame is not
available.
CONVENIENCE SAMPLING
• It is also known as accidental, haphazard or accessibility
sampling.
• The sample is selected from elements of a population that are
easily accessible.
• A readily available group of individuals is used. MERITS
• Practical approach
DEMERITS • Relies on readily
• Opportunistic and available units
voluntary nature of • Administrative
participants convenience
• Lack of • Ease of access
representativeness
JUDGMENTAL SAMPLING
Purposive Sampling
The researcher purposively selects those individuals who are
representative of the population.
MERITS
The investigator selects the sample on the basis on the basis of (i) It is less costly and more
his/her own judgment. readily accessible to the
researcher.
DEMERITS (ii) Purposive sampling
(i) introduces subjectivity in guarantees that those
the sampling individuals will be included
(ii) the ability of the in the sample that are
investigator to assess which representative of the
element is typical of the population.
population.
(iii) the inferential statistics
can’t be used legitimately,
there is an assumption of
randomness.
QUOTA SAMPLING
It is a non-probability sampling technique where the assembled
sample has the same proportions of individuals as the entire Merits
population with respect to known characteristics, traits or focused •Moderate cost
phenomenon. •Very extensively used and understood
•No need for the list of population
Two types of quota elements
1.Equitable quota
An equal number of sample is selected. Demerits
Example •Variability and bias cannot be measured
100 samples are selected all from 4 areas, i.e A, and controlled
B,C,D •Projecting data beyond the sample not
justified
2.Ratio based quota
On the basis of ratio, sample units are selected from
all areas of the population.
Example
If we have to select 200 samples from country X
then we will select 45% from A, 30% from B, 15%
from C and 10% from D.
SNOWBALL SAMPLING: chain sampling
It relies on referrals from initial subjects to generate additional
subjects
When participants are hard to find, for example, a study
investigation on cheating in exams

Merits
•The ability to recruit hidden populations.
•Economical

Demerits
•It can lead to bias
•It’s not possible to determine the sampling error
•There is no guarantee about the
representativeness of samples
 PROBABILITY SAMPLING

It is from a homogeneous
population.

Each member of the population has an Ensures Time taking

equal chance of as a sample unit Minimum Costly
biasness Requires
Size of the population must
population
be known.
frame
Every element has a known nonzero
probability of being sampled.

Involves random selection at some point.

SIMPLE OR UNRESTRICTED RANDOM SAMPLING (1)
• This technique is applied in a homogeneous
population.
• The purest form of probability sampling.
ADVANTAGES
• Assures each element in the population has an
• minimal knowledge of population
equal chance of being included in the sample.
needed
• Random number generators
Sample Size • External validity high; internal
Probability of Selection = Population Size validity high; statistical estimation
of error
• Easy to analyze data
DISADVANTAGES
• High cost; low frequency of use
• Requires sampling frame
• Does not use researchers’ expertise
• Larger risk of random error than stratified
SIMPLE OR UNRESTRICTED RANDOM SAMPLING (2)
Methods Adopted :
Lottery Method Step:
Constructing the Sampling
Frame

Writing numbers listed in the sampling

frame on small pieces of papers and
placing them in a vessel

Mixing all papers well and drawing

out one piece of paper at a time
from the vessel.
 SYSTEMATIC SAMPLING (1)
• An initial starting point is selected by a random
process, and then every nth number on the list is
selected
• n=sampling interval
o The number of population elements between the
units selected for the sample
o Error: periodicity- the original list has a systematic
◼Advantages
pattern
◼Moderate cost; moderate usage
Sampling fraction ◼External validity high; internal validity
The ratio between sample size and population high; statistical estimation of error
(K)= N/n
◼Simple to draw sample; easy to verify
N =size of population
n = desired sample size ◼Disadvantages

◼Periodic ordering

◼Requires sampling frame

 07 selecting our sample

06 Selecting the first unit

05 Calculating the sampling fraction

How to draw a 04 Assigning numbers to cases

SYSTEMATIC
03 Listing the population
RANDOM
SAMPLE? 02 Choosing your sample size
(2)
01 Defining the population
STRATIFIED SAMPLING (1)
• Sub-samples are randomly drawn from samples within
different strata that are more or less equal on some
◼Advantages
characteristic
◼Assures representation of all groups
• Why? in sample population needed
o Can reduce random error ◼Characteristics of each stratum can
o More accurately reflect the population by more be estimated and comparisons made
proportional representation ◼Reduces variability from systematic
• How?
◼Disadvantages
1. Identify variable(s) as an efficient basis for
◼Requires accurate information on
stratification. Must be known to be related to
dependent variable. Usually a categorical variable proportions of each stratum
2. Complete list of population elements must be ◼Stratified lists costly to prepare
obtained
3. Use randomization to take a simple random sample
from each stratum
STRATIFIED SAMPLING (2)

Types of Stratified Samples

• Proportional Stratified Sample:

The number of sampling units drawn from each
stratum is in proportion to the relative population size
of that stratum

• Disproportional Stratified Sample:

The number of sampling units drawn from each stratum is allocate
ed according to analytical considerations e.g. as variability increases
sample size of stratum should increase
CLUSTER SAMPLING (1)
• The primary sampling unit is not the individual element,
but a large cluster of elements. Either the cluster is
randomly selected or the elements within are randomly
selected Types of Cluster Samples
• Area sample:
• Why?
Primary sampling unit is a
• Frequently used when no list of population geographical area
available or because of cost
• Ask: is the cluster as heterogeneous as the • Multistage area sample:
population? Can we assume it is representative? Involves a combination of two or
more types of probability
sampling techniques. Typically,
progressively smaller
geographical areas are randomly
selected in a series of steps
CLUSTER SAMPLING (2)
STRATIFICATION VS CLUSTERING
STRATIFICATION CLUSTERING
All strata are Only a subset of
represented in the clusters are
sample. represented in the
sample.
Less error compared More error
to simple random. compared to simple
random.
More expensive to Reduces cost to
obtain stratification sample only some
information before areas or
sampling. organizations.
MULTISTAGE SAMPLING
 Sampling process carried out in various stages.
 An effective strategy because it banks on multiple
randomizations.
COUNTIES 01
 Used frequently when a complete list of all members of
the population does not exist and is inappropriate.
SEGMENTS 02
 Merits
• Introduces flexibility in the sampling method.
• Enables existing divisions and sub divisions of
population to be used as units.
HOUSEHOLDS 03
• Large area can be covered.
• Valuable in under developed areas.
INDIVIDUALS 04
 Demerits
Less accurate than a sample chosen by a single stage
process.
MULTIPHASE SAMPLING
 Used for studies to be carried out in
multiple phases. 1: ALL SAMPLES
 For e.g. A cross - sectional study on
nutrition may be carried out in phases
2: SUBSAMPLE
➢ Phase-1: K.A.P. study in all families
➢ Phase -2: Dietary assessment in
subsample 3: SUBSAMPLE COVERED
➢ Phase-3:anthropometric examination in IN 2ND PHASE
sub- sample of family members
covered in 2nd phase
 ERRORS SAMPLING
ERRORS
SAMPLING NON SAMPLING
ERRORS ERRORS NON SAMPLING ERRORS
➢ Data specification inadequate &
SAMPLE SAMPLE AND inconsistent with respect to objective of
CENSUS census.
➢ Inaccurate or inappropriate methods of
interview, observation, definitions.
➢ Lack of trained & experienced
SAMPLING ERRORS investigators.
BIASED ELIMINATION OF ➢ Errors due to non response.
ALL SOURCE OF ➢ Errors in data processing operations • Errors
SAMPLING BIAS committed during presentation.
ERRORS
UNBIASED INCREASE
SAMPLE SIZE
DETERMINING SAMPLE SIZE
What data do you need to consider
• Variance or heterogeneity of population
➢ Previous studies? Industry expectations? Pilot
study?
➢ Sequential sampling
➢ Rule of thumb: the value of standard
deviation is expected to be 1/6 of the range. Formulas:
• The degree of acceptable error (confidence Means n = (ZS/E) 2
interval) Proportions n = Z2 pq/ E2
• Confidence level Percentiles n = pc (100 – pc) Z2/ E2
• Generally, we need to make judgments on all
these variables Z at 95% confidence = 1.96
Z at 99% confidence = 2.58
https://www.slideshare.net/7muku
t/sampling-techniques-49115431,
accessed June 26th, 2020 http://123phnotes.blogspot.co
https://www.slideshare.net/drbhar m/2018/10/sampling.html
atpaul/sampling-techniques- https://medsdrive.com/samplin
48927352, accessed June 26th, g/, accessed June 26th, 2020
2020 Dr. Moataza Mahmoud Abdel
https://www.slideshare.net/shaya Wahab, Sampling technique
a/research-methodology-insaniah, and sample size, Lecturer of
Biostatistics, High Institute of
accessed June 25th, 2020 Public Health University of
https://www.slideshare.net/aaaad Alexandria
itantia/2-presentations-
malhotraorgnlppt11, accessed
June 25th, 2020
THANK YOU