87

ASEAN Journal for Science Education 2(2) (2023) 87-94

ASEAN Journal for Science

Education
z
Journal homepage: [Link]

Review of Sampling Techniques for Education

Moses Adeleke Adeoye

Al-Hikmah University, Ilorin, Nigeria

Correspondence: E-mail: princeadelekm@[Link]

ABSTRACT ARTICLE INFO

Article History:
This paper offers a thorough explanation of the procedure Submitted/Received 02 Jan 2023
First Revised 18 Feb 2023
for aspiring authors to learn more about data-gathering Accepted 25 Apr 2023
techniques and the application of sampling strategies in First Available online 27 Apr 2023
completing papers for publication in a range of academic Publication Date 01 Sep 2023

fields. The sections describe the data-gathering methods ____________________

Keyword:
used & sampling techniques to clarify those differences. Both Non-probability sampling,
probability sampling and non-probability sampling are Probability sampling,
employed in statistics. The advantage of probability sampling Sampling.

ensures the sample will be represented in the population.

Simple random, systematic, stratified, and cluster sampling
are some of the different types of probability sampling
techniques. Non-probability sampling chooses the sample
based on individual judgment. Non-probability sampling
technique is considered the best because it includes all
subjects’ representation of the entire population.
Convenience sampling, sequential sampling, quota sampling,
judgmental sampling, and snowball sampling are just a few
examples of non-probability sampling. The differences
between non-probability sampling techniques and
probability sampling approaches are examined.

© 2023 Bumi Publikasi Nusantara

 Adeoye, Review of Sampling Techniques for Education | 88

1. INTRODUCTION

The core goal of data collection techniques includes both secondary and primary data
gathering that is focused on the kinds of qualitative and quantitative data discussed in the
previous section. The systems for collecting data have been developed and the basic data-
collecting techniques are the main sources. Field observations, interviews, and casual chats
are examples of qualitative sources whereas survey questions and interview topics are
examples of quantitative data sources (Peshkin, 2000). The details of the data's main source
acquisition are provided in the following sections. Secondary data gathering techniques are
data that was gathered by a party other than the user (Smith, 2008).
This data source provides information about the current state-of-the-art method (Hsu,
2015). The supplemental data sources may include both internal and external information
sources and cover a wide range of topics.
2. METHODS

This paper is a literature survey, taking data from internet sources especially articles from
international journals. Data were obtained, collected, and summarized to build this paper.
3. RESULTS AND DISCUSSION
3.1. Sampling
Sampling is the process of employing a portion of a population to represent the entire
population in survey research. There are some instances where data sampling techniques
help in clarifying (Afifah et al., 2022). The sampling employs a smaller number of people with
representative traits to represent the entire large scale with more realistic costs and
timelines. You must choose who will be included in your sample list and how to pick those
who will best reflect the entire population. Sampling is a method that you might use to
determine how to approach things. Sampling methodology is a statistical approach used by
researchers in acquiring data and examining the population. Numerous sampling methods
can be used and they can be split into two categories. All of these sampling techniques may
aim for difficult-to-reach groups (Wahab et al., 2023).
Sampling vastly expands the opportunities for research because it enables you to study
larger target populations with the same resources as you would for smaller ones. Using
sampling can help make your research less constrained by costs, delays, and complexity
associated with various population sizes. To obtain pertinent data from the population many
sampling approaches are available in statistics such as Probability Sampling and Non-
probability Sampling.
3.2. Probability sampling
Probability sampling also referred to as random sampling is a type of sample selection
where randomization is used rather than deliberate choice. The probability sampling method
makes use of a random selection process. In this method, each eligible person has a chance
to choose a sample from the entire sample space. The advantage of probability sampling
ensures the sample will be representative of the population. The various forms of probability
sampling techniques include simple random, systematic, stratified, and clustered. A detailed
discussion of the various probability sampling techniques is illustrated with examples as
follows:

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
89 | ASEAN Journal for Science Education, Volume 2 Issue 2, September 2023 Hal 87-94

(i) Simple Random Sampling. Every member of the population has an equal probability of
being chosen for the sample when using simple random sampling. It resembles drawing
a name out of a hat in several ways. By making the population anonymous, each object
or person in the population is a number and then selecting those numbers at random
simple random sampling can be carried out. Simple random sampling eliminates any
bias risks from the sampling process and is quick, easy, and inexpensive to use. It also
provides the researcher with no means of control which increases the likelihood that
unrepresentative groupings will be chosen at random. Every item in the population has
an equal and likely probability of being chosen for the sample when using a
straightforward random sampling procedure. This method is referred to as the "Method
of chance Selection" since the decision to select an item is entirely based on chance. It
is referred to as "Representative Sampling" because of the size of the sample and the
random selection of the item. For instance, let's say we wish to choose 200 kids at
random from a school. We can give each student in the database of school a number
between 1 and 500 and use a random number generator to choose a sample of 200
numbers (Agustina and Nandiyanto, 2021). The benefits of using this method are
simplicity and impartiality. Although it is straightforward to carry out, simple random
sampling isn't employed very frequently due to the subgroup of the bigger group being
drawn at random from the large population, any traces of bias are eliminated
(Mohamad and Masek, 2021). The outcome might be reasonably dependable because
it is much less complicated. When conducting larger populations, it takes little to no
specialized understanding. The inherent shortcomings of a simple random sample
include the time required to compile the entire list of a particular community. Difficulty
gaining access to population-wide lists. impose protracted retrieval procedures that
may hinder a researcher's capacity to gather the most precise data on the complete
population set. It causes implicit bias, which leads to a skewed study.
(ii) Systematic Sampling. Systematic sampling also known as systematic clustering is a sort
of probability sampling technique in which sample participants from a broader
population are chosen using a predetermined, periodic interval but a random beginning
point. The population size is multiplied by the required sample size to determine the
interval which is known as the sampling interval. According to the rule, the Nth object
or person is chosen even though the picks are made at random, the researcher can alter
the interval at which they are chosen by ensuring that no selections are unintentionally
grouped. After a predetermined sample interval, the items are chosen at random from
the target population and then the other ways are chosen. To compute it, the required
population size is by the total population size. Consider the following scenario: 300
students' names are arranged in a school in reverse alphabetical order. For example, 15
students were randomly selected while the 15th participant will be chosen at random
from the sorted list starting with number 5. We might conclude by presenting a sample
of a few students. With few resources, systematic sampling's advantages are relatively
simple to plan, carry out, evaluate, and comprehend. It gives researchers a certain
amount of control and rigor to fit particular requirements. Systematic sampling
eliminates clustered selection. The drawbacks of systematic sampling presuppose that
the population's size can be approximatively determined. It shows a naturally occurring
level of unpredictability along the selected metric because researchers might be able to
design their systems to maximize the likelihood of reaching a specified result rather

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
 Adeoye, Review of Sampling Techniques for Education | 90

than letting the random data create a representative answer, there is a higher risk of
data manipulation.
(iii) Stratified Sampling. Random selection is used in stratified sampling within
predetermined groupings. Knowing information about the target population helps
researchers stratify it (subdivide it) in a way that makes sense for the research. It raises
the issue of how to stratify a population which increases the possibility of bias. To finish
the sampling process, a stratified sampling approach divides the entire population into
smaller groups. A few features of the population are used to establish the tiny group.
The statisticians choose the sample by drawing at random from the smaller groups
created by the population division. For instance, there are three bags (A, B, and C)
containing various eggs. There are 50 eggs in bag A, 100 eggs in bag B, and 200 eggs in
bag C. We must randomly select a sample of eggs from each bag. Let's say there are 20
eggs in bag C, 10 eggs in bag B and 5 eggs in bag A. The benefit of stratified sampling is
that it allows you to get reasonably accurate estimates for all subgroups connected to
your research subject. The accuracy increases the statistical power of your analysis to
identify group differences. The population is divided, which lowers survey expenses and
streamlines data collecting. It aids in maintaining the full range of the population's
diversity inside the sample. The drawback of these datasets is the complexity of their
analysis. Any relatively similar strata will be unsuccessful if researchers cannot develop
appropriate strata.
(iv) Clustered Sampling. Cluster sampling is a probability sampling technique in which you
divide a population into clusters and then choose a random subset of these clusters as
your sample. Researchers analyzed a sample that consists of multiple sample
parameters such as demographics, background, or other attributes. Instead of selecting
the entire population, cluster sampling allows the researchers to collect data by
bifurcating the data into small, more productive groups. Cluster sampling chooses
individuals from the target population at random. Simple random sampling is used in
this method to sample the cluster of the population. The advantage was that the
sampling procedure needed fewer resources. Therefore, administrative and travel costs
are typically less expensive. The feasibility of the sample is increased when the complete
population is divided into homogeneous groups. Cluster sampling has drawbacks since
it might lead to biases in representing the full population and drawing conclusions about
the entire population. Compared to samples created using other sampling methods, the
cluster approach is more susceptible to sampling error.
3.3. Non-probability sampling
Non-probability sampling refers to the purposeful selection of objects or subjects for the
sample by the researcher by their objectives or expertise. In contrast to random selection,
non-probability sampling chooses the sample based on individual judgment. Not every
member of the population will be given the chance to participate in the research under this
methodology.
There are many other non-probability sampling techniques, such as convenience sampling,
sequential sampling, quota sampling, judgmental sampling, and snowball sampling. Non-
probability sampling types are as follows:
(i) Convenience Sampling. Convenience sampling involves selecting a sample from a group
of people who are simple to get in touch with. The terms grab sampling and availability
sampling are also used to describe this method of sampling. Convenience sampling
methods are selected from the population directly because they are conveniently

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
91 | ASEAN Journal for Science Education, Volume 2 Issue 2, September 2023 Hal 87-94

available for the researcher. The samples are easy to select and the researcher did not
choose a sample that outlines the entire population. This method of data collection is
practical. The sample does not accurately reflect all participants depending on their
availability and accessibility.
Although a large amount of bias might be introduced if it is conducted early or as a first
step. The advantages of convenience sampling are that is cheap, efficient, and simple
to implement. It is useful as an intervention to correct dissatisfaction. It offers a way to
receive specific feedback from an individual’s demographic profile than generalizing. It
is a lot easier to analyze the info instead of worrying about participant selection and
interviews. It allows the completion of multiple studies simultaneously since one
participant can provide data on multiple subjects in just a few minutes. The
disadvantage of convenience sampling is that the sample lacks clear generalizability. It
doesn’t provide a representative result. It is easier to provide false data if the researcher
determines the exclusion of that data will adversely influence the results. it can lead to
decision-making circumstances that lead people and organizations in the wrong
direction. It is challenging to replicate the results.
(ii) Consecutive Sampling. Consecutive sampling is described as a non-probability sampling
strategy where samples are chosen at the researcher's convenience, similar to
convenience sampling but with a small difference. Consecutive sampling is similar to
convenience sampling with a small difference. The researcher selects an individual or a
group of persons for sampling. The researcher then does additional research for some
time, analyzes the findings, and, if necessary, switches to a different group. It provides
a way for researchers to improve the representativeness of their samples in the target
population. The advantage of consecutive sampling is quick and easy to carry out.
The researcher selects individuals consecutively until they reach the number required
for the study. It allows researchers to choose their sample to improve the reliability and
validity of individuals. The researcher has many options when it comes to sampling size
and sampling schedule. To produce definitive results, the sample plan must be based
on the nature of the research. Minor changes and adjustments can be made right at the
beginning of the research to avoid considering research bias. The disadvantage of
consecutive sampling is that selection of participants is not random, people in that
setting must also be representative of the desired population. It is uncommon to
identify potential participants because doing so could skew the results.
(iii) Quota Sampling. Quota sampling is a non-probability sampling technique where
researchers compile a convenience sample of participants who accurately reflect the
population. Researchers select these people based on particular characteristics while
the final subset will only be chosen based on the researcher's familiarity with the group.
By designating which individuals or groups should be recruited for a survey by particular
groupings, the strategy seeks to achieve a dispersion over the target population. For
instance, your quota can call for a specific proportion of men and women. As an
alternative, you might want your samples to fall within a particular age range, income
range, or ethnic group. Bias may be introduced during the selection process; for
instance, volunteer bias may skew the sample in favor of those with spare time and an
interest in participating. Creating individuals with similar characteristics or traits
relevant to the research will ensure participants in the study belong to only one group.
One specific city is being studied to determine the percentage of subgroups in quota
sampling. When choosing the sample size, at the very least, pick 100–200 people from

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
 Adeoye, Review of Sampling Techniques for Education | 92

each sampling category. The research group's volunteers determine how accurate the
study is.
The advantages of quota sampling are quite easy to conduct and administer as
compared to other similar research sampling methods. It is suitable for research where
the researcher has a time limit to conduct the study. It saves a lot of time for the
researcher. It can be used for research with a limited budget. Quota sampling helps in
an easy comparison of two groups of research. Quota sampling is beneficial in scenarios
where researchers have specific criteria to conduct research. The disadvantage of quota
sampling does not allow a random selection of participants. It increases the risk of being
biased as a researcher in approach. It lacks randomness in the selection of participants
as it is affected by many factors.
(iv) Purposive or Judgmental Sampling. Purposive sampling often referred to as
authoritative or judgmental sampling is a type of non-probability sampling in which
researchers use their judgment to select people from the public to take part in their
surveys. Since the researcher's experience was essential in producing the sample, their
knowledge is used to choose the samples. Participants in the sample are purposefully
chosen by researchers based on their familiarity and grasp of the pertinent study topic.
It is a quick and comparably easy way to compile a variety of results or responses to
create a representative sample. There are several purposive sampling types that
researchers can use such as heterogeneous variation, typical case sampling, critical case
sampling, or total population.
The advantages of Purposive Sampling are the opportunity to create justification to
generalize from the sample. A critical sample can be used in determining the value of
an investigation while the sampling approach allows an in-depth analysis of the
information. The flexibility of purposive sampling allows researchers to save time and
money while they are collecting data. It helps researchers to develop insight as a key
point in observing a strong perspective that represents the general public.
It helps researchers to identify the extreme perspectives that are present in each
population group. Researchers achieve a lower margin of error using the purposive
sampling approach because the information they collect comes straight from the
source. The disadvantages of purposive sampling allow you to generalize from specific
samples to a larger population group making statements about the validity or accuracy
of your discoveries complex. It can provide roadblocks in the way of a final result. It may
be challenging to defend the representative nature of a sample. When used on sizable
populations, it may be ineffectual. It can be virtually impossible to determine if there is
a sampling error in the present information (Shaturaev, 2022).
(v) Snowball Sampling. Snowball sampling is a non-probability sampling technique that
involves enlisting new units as sample members from among the existing ones. It can
be used to study people with particular traits who might be hard to find in other ways.
The participation spreads like a snowball tumbling downhill across a network of
connected people.
This approach can be useful if the researcher is unfamiliar with the target demographic
and has difficulty reaching out to or accessing them. Snowball sampling has the benefit
of being quick and simple to discover subjects because they originate from trustworthy
sources. It is convenient and not expensive as compared to other methods. It helps in
situations where they ask for a reference from people known to each other. The
disadvantages of snowball sampling might only be able to reach out to a small group of
people and may not be able to complete conclusive results.

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
93 | ASEAN Journal for Science Education, Volume 2 Issue 2, September 2023 Hal 87-94

3.4. Probability sampling vs non-probability sampling methods

To separate the sampling techniques, the many differences between probability sampling
and non-probability sample approaches have been investigated (Table 1).
Table 1. Differences between probability sampling and non-probability sample
No Probability Sampling Methods Non-Probability Sampling Methods
1 The probability sampling Non-probability sampling is chosen by the researcher
methodology is based on the theory based on their judgment rather than by random selection
of selecting from a larger population
2 Random sampling method Non-random sampling method
3 The population is selected randomly The population is selected illogically
4 They employed a conclusive study The study is preliminary
5 It takes a while to gather the data These are quick and simple techniques to gather data
6 The sample uses a process to secure The depiction of the population's demographics is
the population's demographics typically distorted because the sampling methodology is
random
7 The selection criteria are defined by The sample nor the criteria for selecting it are vague and
the research design before the rapid
market research study starts
8 This form of sampling is completely The results of this form of sampling are completely
objective, the conclusions drawn are biased, which makes the research speculative
also objective and conclusive
9 There is a guiding hypothesis before The non-probability sampling is developed after the
the investigation and the goal of the research hypothesis is completely investigated
probability sampling approach is to
support that hypothesis

4. CONCLUSION

This essay offers a discussion on the need for high-quality educational research. It
exemplifies how the assessment of the quality of educational research satisfies both the
practical demands for relevance to modern social science and the epistemic standards for
rigor measurement and statistical models. Probability concepts provide the framework for
determining if apparent discrepancies in measures are likely a random occurrence given the
underlying variability present in the real world.
5. AUTHORS’ NOTE

The authors declare that there is no conflict of interest regarding the publication of this
article. Authors confirmed that the paper was free of plagiarism.
6. REFERENCES

Afifah, S., Mudzakir, A., and Nandiyanto, A. B. D. (2022). How to calculate paired sample t-
test using SPSS software: From step-by-step processing for users to the practical
examples in the analysis of the effect of application anti-fire bamboo teaching materials
on student learning outcomes. Indonesian Journal of Teaching in Science, 2(1), 81-92.

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295
 Adeoye, Review of Sampling Techniques for Education | 94

Agustina, S., and Nandiyanto, A. B. D. (2021). The effectiveness of distance learning using
learning management system media and whatsapp groups at senior high school.
Indonesian Journal of Multidiciplinary Research, 1(1), 89-98.
Hsu, T. C. (2015). Research methods and data analysis procedures used by educational
researchers. International Journal of Research and Method in Education, 28(2), 109-133.
Mohamad, N., and Masek, A. (2021). Modes of facilitator skills for group learning among
design and technology teachers in secondary schools. Indonesian Journal of Teaching in
Science, 1(1), 17-20.
Peshkin, A. (2000). The nature of interpretation in qualitative research. Educational
Researcher, 29(9), 5–9.
Shaturaev, J. (2022). Assessment methodology role of family entrepreneurship in the sphere
of services. ASEAN Journal of Economic and Economic Education, 1(2), 67-78.
Smith, E. (2008). Pitfalls and promises: The use of secondary data analysis in educational
research. British Journal of Educational Studies, 56(3), 323-339.
Wahab, N.A., Mahmood, N.H.N., Minghat, A.D. (2023). Science education research
methodology: A case study investigating the correlation between construction, safety,
accident, and the effectiveness construction industry development board (CIDB) green
card training program. ASEAN Journal of Science Education, 2(1), 7-16.

DOI:
p- ISSN 2828-3287 e- ISSN 2828-3295