Chapter 5
Sampling
Lecturer: Le Hoai Kieu Giang
Email: lhkgiang@[Link]
Contents:
- Sampling methods
- Sample size
2
1. Definition
Sampling (việc lấy mẫu) is selecting some of the elements in a
population.
A sample (mẫu): A subgroup of the elements of the population selected
for participation in the study
An element (phần tử): the individual participant or object on which the
measurement is taken. (E.g. a person, a representative of an
organization/ a household,…)
A population (tổng thể/ đám đông): The aggregate of all the elements,
sharing some common set of characteristics, that comprises the universe
for the purpose of the research problem. 3
1. Definition
Study population (Tổng thể/ đám đông nghiên cứu): All elements can
be identified and approached for sampling in practice.
A sampling unit (đơn vị lấy mẫu) is an element, or a unit containing the
element.
A sampling frame (khung mẫu) consists of a list or set of directions
for identifying the target population. Examples of a sampling frame
include the telephone book, an association directory listing the firms in an
industry, a mailing list purchased from a commercial organization, a city
directory, etc.
4
2. Why Sample?
• Lower cost: The economic advantages of taking a sample rather than the whole
population are massive.
• Greater accuracy of results: Sampling possesses the possibility of better
interviewing (testing), more thorough investigation of missing, wrong, or
suspicious information, better supervision, and better processing.
• Greater speed of data collection: save time
• It is necessary for the research that leads to destruction or changes in object
properties.
E.g.: Crash-test cars
5
3. Sampling process
Identify the study population and elements
Identify the sampling frame
Determine the sample size
Determine the sampling method
Sampling
6
4. Sampling methods
Probability Nonprobability
(Chọn mẫu xs) (Chọn mẫu phi xs)
7
4. Sampling methods
Probability
(Chọn mẫu xs)
• Know in advance the probability of the occurrence of elements in the
sample.
• The sampling process follows mathematical rules and can not be
changed.
• Only probability samples offer the opportunity to generalize the findings
to the population from the sample population.
8
4. Sampling methods
Nonprobability
(Chọn mẫu phi xs)
• Nonprobability sampling is arbitrary and subjective. (meaning “as
they wish” or “wherever they find them”).
• Each element of the population does not have a known chance of
being included.
• Generalizing the findings to the population from the sample
population may not reliable.
9
10
4. Sampling methods
Probability
(Chọn mẫu xs)
• Each population element has a known
and equal chance of selection.
Simple random
(ngẫu nhiên đơn giản) • The sample is drawn by a random
procedure from a sampling frame.
• It requires a sampling frame, can be
time-consuming and expensive. è It
is impossible for a large population.
12
4. Sampling methods
Probability
(Chọn mẫu xs) • Randomly select a starting point,
based on a sampling interval (k), to
Systematic Sampling determine the next elements from the
(Hệ thống) sample frame.
k = Skip interval = Population size/ Sample size
• This is a more commonly used
method than the simple random
method.
13
4. Sampling methods
Probability
(Chọn mẫu xs)
Disadvantage: A concern with
systematic sampling is the
Systematic Sampling
possible periodicity in the
(Hệ thống)
population (the frequency of
periodicity is equal to the skip
interval) è oversampling of a type
of element
14
4. Sampling methods
Probability
(Chọn mẫu xs)
Some ways to avoid such bias include:
Systematic Sampling
(Hệ thống) • Randomize the population before
sampling
• Change the random start several times
in the sampling process.
15
4. Sampling methods
Probability • The elements within a stratum should be as
homogeneous as possible, but the
(Chọn mẫu xs) elements in different strata should be as
heterogeneous as possible. (Cùng tầng
Stratified Sampling đồng nhất, khác tầng dị biệt)
(Phân tầng)
• Elements are selected from each stratum
by a random procedure.
• Stratified sampling can ensure that all the
important subpopulations are represented
in the sample. This is particularly important
if the distribution of the characteristic of
interest in the population is skewed.
16
4. Sampling methods
Probability
(Chọn mẫu xs) • In proportionate stratified sampling,
each stratum is properly represented
Stratified Sampling so that the sample size drawn from the
(Phân tầng) stratum is proportionate to the stratum’s
share of the total population.
• Any stratification that departs from the
proportionate relationship is
disproportionate stratified sampling.
17
4. Sampling methods
Probability
(Chọn mẫu xs) • Heterogeneity within subgroups,
and homogeneity between subgroups
Cluster Sampling (cùng nhóm dị biệt, khác nhóm đồng
(Theo nhóm) nhất)
• Randomly choose several subgroups.
• Area sampling: the clusters are
determined based on geographic
criteria.
18
4. Sampling methods
19
4. Sampling methods
20
4. Sampling methods
Nonprobability • Choose elements based on
(Chọn mẫu phi xs) convenience, ease of access, and ease
of obtaining information.
Convenience Sampling • Respondents are selected because
(Thuận tiện) they happen to be in the right place at
the right time.
• It is the least expensive and least time-
consuming of all sampling techniques.
• The findings of sample can not be
generalized to a population parameter.
21
4. Sampling methods
Nonprobability • It is a form of convenience sampling in
(Chọn mẫu phi xs) which the population elements are
selected based on the judgment of the
researcher.
Judgment sampling
(Phán đoán) • Judgmental sampling is low cost,
convenient, and quick, yet it does not
allow direct generalizations to a specific
population.
• Judgmental sampling is subjective and
its value depends entirely on the
researcher’s judgment, expertise.
22
4. Sampling methods
Nonprobability
(Chọn mẫu phi xs) • The relevant control characteristics,
which may include sex, age, and
income, type of organization,… are
Quota sampling identified on the basis of judgment.
(Theo lớp)
• The quotas are assigned so that the
proportion of the sample elements
possessing the control characteristics
is the same as the proportion of
population elements with these
characteristics.
23
4. Sampling methods
Nonprobability
(Chọn mẫu phi xs) • An initial group of respondents is
selected randomly.
Snowball sampling • After being interviewed, these
(Theo mầm) respondents are asked to identify
others who belong to the target
population.
• Snowball sampling is often used when
the population is small (rare
characteristic) and hard to realize.
24
5. Sample size
Sample size refers to the number of elements to be included in the study.
Determining the sample size is complex and involves several qualitative and
quantitative approaches as well as data analysis methods considerations.
25
5. Sample size
Some common ways to determine sample size:
Ø Exploratory Factor Analysis (EFA):
• At least: N = 50 (for preliminary research)
• A more suitable sample size would have a ratio of 5:1 or 10:1 compared to the
number of observations (the items in the measurement scale).
Ø Multiple Regression Analysis:
N ≥ 50 + 8p
p: the number of independent variables
(Thọ, N. Đ (2011). Phương pháp NCKH trong Kinh doanh; Hair et al. (2019). Multivariate Data Analysis)
26
5. Sample size
Ø Qualitative research:
Purposeful sampling (or theoretical sampling: chọn mẫu lý thuyết): It
involves selecting participants, cases, or contexts based on their
potential to contribute to the development and refinement of
theoretical concepts.
The sample size is often defined based on saturation point.
(Thọ, N. Đ (2011). Phương pháp NCKH trong Kinh doanh)
27
5. Sample size
Rules of thumb:
• 30 < Sample sizes < 500 are appropriate for most research
• Where samples are to be broken into subsamples (males/females,
juniors/seniors, …), a minimum size of 30 for each subgroup is
necessary
• In multivariate analysis, the sample size should be several times
(preferably 10 times or more) as large as the number of variables in the
analysis.
(Roscoe, 1975) 28