International Journal of Academic Research in Education (IJARE)

Volume 8, Issue 1, ss. 75-84, December 2022

Sampling techniques involving human subjects: Applications, pitfalls, and suggestions for
further studies

Feyisa Mulisa 1
Ambo University, Institute of Education and Behavioral Science, Ambo, Ethiopia

Abstract: The most commonly used sampling techniques in systematic investigations are probability and
nonprobability methods. While probability sampling is based on the principle of a random selection of
participants in a particular study, non-random selection is the basis of probability sampling. The random and
non-random classifications appear to have some potential flaws and are insufficient to represent all sampling
procedures involving human participants. Similarly, most authors believe that they use random sampling
techniques, although, in reality, they do not use true random sampling. Therefore, the objective of this article is
to highlight that sampling techniques can be characterized as true-random, quasi-random, or nonrandom, rather
than merely random and non-random. Attempts have been made to demonstrate how inadequate random and
non-random sampling classification is, the characteristics of true-random, quasi-random, and nonrandom
sampling procedures, and when each sampling procedure is appropriate. How it is conceivable to estimate the
characteristics of the population directly or indirectly from the sample has been addressed in light of the selection
of sampling units technique.

Keywords: True-random; quasi-random; non-random; sampling techniques

Suggested Citation: Mulisa, F. (2022). Sampling techniques involving human subjects: Applications, pitfalls,
and suggestions for further studies. International Journal of Academic Research in Education, 8(1), 75-84.
DOI:10.17985/ijare.1225214

Article History: Submitted 27 December 2022; Revised 29 December 2022; Accepted 30 December 2022

INTRODUCTION

Sampling is a scientific method used in systematic studies to select units from a target population to represent the
entire study population (Berndt, 2020; Singh & Masuku, 2014; Taherdoost, 2016). Researchers usually rely on
sampling to estimate the characteristics of the population by studying sample characteristics (Walters, 2021).
When researchers select a sample for a study, their primary goal is to understand the features of the entire
population from which the sample was drawn, rather than the sample itself. Sampling is essential for a variety of
reasons, including cost-effectiveness, increased precision, and efficient use of time, particularly for a large
population (Kadam & Bhalerao, 2010; Martinez-mesa et al., 2016). Therefore, to accurately represent the entire
study population, the sample should contain all of the features of the population from which it was drawn.
However, this is not always the case because all the approaches to sampling do not guarantee the representation
of the sample. In systematic research, there are two widely used approaches to sampling: random and nonrandom
(Berndt, 2020; Elfil & Negida, 2017; Khaldi, 2017; Singh & Masuku, 2014).

The term "random sampling" refers to a sampling technique where the likelihood of selecting each unit is the same.
The sample that is selected at random is thought to be fair and representative of the entire population (Singh &
Masuku, 2014). The reduction of sampling error is the main objective of this sampling technique. Non-random
sampling, on the other hand, is a sampling technique in which the sample is chosen based on a specific reason

1
Corresponding Author: fayisamu@gmail.com

2022 IJARE e-ISSN: 2149-2913 / http://dergipark. gov.tr/ijare. All rights reserved.

Mulisa, F. 76

rather than just chance (Martinez-mesa et al., 2016). Here, the sample will be chosen depending on the
researcher's convenience, expertise, or judgment. Bias is seen as a significant limitation of the sample collection
process in non-random sampling. The classification of sampling techniques into random and non-random is
common in virtually all types of literature related to sampling. Nevertheless, there are several potential pitfalls and
issues associated with this classification. This paper, therefore, argues that sampling techniques do not necessarily
fall within the category of random and random sampling. Instead, attention should be given to the third type of
sampling, called quasi-sampling, which is hardly recognized in statistics and research methodologies. This paper
may contribute to the social sciences discipline by enhancing the skills of investigating social phenomena through
the advancement of sampling techniques.

While probability sampling is based on the random selection of participants to a study, nonprobability sampling is
based on the nonrandom assignment of participants to a study (Ary et al., 2010; Martnez-mesa et al., 2016;
Palinkas et al., 2015; Taherdoost, 2016). When it comes to probability sampling, in particular, two key assumptions
often underpin its operation. The assumptions are that each unit/person in the population has a non-zero chance
of being chosen and that one unit/person’s selection is unrelated to the selection of another (Cohen et al., 2018;
Singh & Masuku, 2014). The random selection of participants is based on the assumption that all members of a
particular population have an equal chance of being selected (Elfil & Negida, 2017; Kadam & Bhalerao, 2010).
When using true-random sampling techniques, because the selection procedure is by chance, each unit of the
population from which the sample is drawn has an equal chance of being included in the study. The fundamental
principle underlying the use of random sampling techniques is to ensure that the sample is representative of the
population it is supposed to represent (Elfil & Negida, 2017; Khaldi, 2017; Sargeant, 2012). Random sampling is
likewise important to reduce sampling bias and increase the accuracy of the study (Palinkas et al., 2015; Walters,
2021). This approach is largely common among quantitative researchers (Delice, 2010). Qualitative research, on
the other hand, focuses on the careful selection of data-rich participants to make the most of limited resources
(Campbell et al., 2020; Palinkas et al., 2015).

As mentioned above, in random sampling, the fundamental principle is that each unit of the population is given
the same opportunity to participate in a particular study (Elfil & Negida, 2017; Martnez-mesa et al., 2016). This
means that samples obtained using these methods often guarantee that each member of the population has an
equal, nonzero chance of being included in the sample. As a result, many authors believe that probability sampling
is more accurate in capturing a population's key characteristics, although this is not always the case (Berndt, 2020).
It should be noted that the criteria that state every member of the population has an equal and nonzero chance
of being selected as a sample is very important and should be highlighted. That is, true-random sampling must
provide equal and non-zero opportunities for all members of the study population, and participants must be
chosen without violating such random selection criteria. However, not all the so-called random sampling
techniques have the potential to produce samples that meet these standards. This could be a reason why Berndt
(2020) argues that probability sampling is not as accurate as the authors believe for a variety of reasons, including
population size and diversity.

To put it another way, there are some approaches to sampling that are typically described as ‘true-
random,’ although they are not necessarily ‘true-random’ but used to assign participants to a study at random.
The ambiguity may endure because the border between random and nonrandom sampling is indistinct, and some
authors believe that the two methods share many characteristics (Etikan & Bala, 2017). Therefore, the random
sampling approach itself should be split into true-random and quasi-random, and the aforementioned sampling
techniques would be better presented as quasi-random rather than true-random.

The attempt to classify the sampling techniques into three is not new and dates back to the first half of the 20th
century. In 1945, for example, Ralph Cassady tried to present sampling techniques such as pseudorandom, pure-
random, and stratified sampling (Cassady, 1945). In non-probability sampling, the sample is chosen under non-

International Journal of Academic Research in Education, 8(1)

Sampling techniques involving human subjects … 77

random circumstances, and not every member of the population has a chance of being included in the sample.
While there are no issues with non-random sampling, the random sample technique needs to be revisited. In
particular, some approaches to sampling that are not genuinely random but are seen to be random should be
identified and addressed differently. The objective of this article is to revisit the classification of random and
nonrandom sampling by (a) demonstrating the distinctions among true-random, quasi-random, and nonrandom
sampling techniques; and (b) establishing when true-random, quasi-random, and nonrandom samples are
appropriate.

True-random sampling techniques

The random and nonrandom classifications of sampling do not appear to be adequate enough to cover all the
sampling techniques. To address this shortfall, classifying sampling as true-random, quasi-random, and non-
random appears meaningful. As a result, the following sections attempt to discuss the distinctions among true-
random, quasi-random, and nonrandom sampling procedures. To get started with true-random sampling, it is a
type of sampling that strictly adheres to the principle of randomization and ensures that every member of the
population has an equal chance of being selected. In other words, researchers must start with the assumptions
that the sampling frame must be readily available, that each member of the population has an equal chance of
being selected without compromising those assumptions, and that the selection procedures must be carried out
rigorously according to the set of criteria. In other words, every member of the population must be listed without
missing data, and each unit must have an equal chance of being selected in the sample. The basic notion behind
this kind of sampling is that a sample selected through this process gives a more accurate population estimate
(Singh & Masuku, 2014). This kind of sampling is typically required when a study is quantitative and the results are
intended to be directly extrapolated to the entire population (Berndt, 2020; Mulisa, 2022; Sargeant, 2012;
Taherdoost, 2016). In particular, while quantitative researchers largely use random sampling, all qualitative
researchers use non-random sampling (Gill, 2020; Palinkas et al., 2015).

To recap, true-random sampling is a sampling approach that uses a complete sampling frame, allows every
member of the population an equal chance to be selected, and proceeds with the selection accordingly without
any compromise. For example, if researchers wish to take a sample of 50 participants from a population of 200,
they can select them using the lottery method. If all 50 participants were exactly selected depending on the lottery
method and no adjustment was made to the selection by the end, the researchers used true-random sampling.
However, the assumption of true-random sampling is violated if certain modifications are made to the sample,
such as replacing nonresponding participants. Ideally, a complete population list must be available, and every n th
person in every population interval becomes a sample unit in systematic sampling. True-random sampling is when
researchers choose every participant in the same way, without making any adjustments, and rigorously follow the
nth principle. It is achieved when each unit in the population has an equal chance of being chosen, as in simple
random sampling, and when the chance of being chosen is known, as in systematic sampling. However, the precise
application of such a randomization principle may not always be feasible for a variety of reasons in social research,
including ethics, accessibility, resources, and participant attrition. This type of sampling, on the other hand, is quite
practical and more relevant among physical scientists than social scientists.

Conducting true-random sampling appears to be quite essential under some circumstances, whereas it may not
be such essential in other circumstances. In particular, it is needed when a sample is a direct representation of the
population. In particular, Bordens & Abbott (2018, p. 168) stated that “when you want to apply your findings
directly to a population,” this kind of sampling is crucial. For example, in a political poll, researchers may survey to
predict the precise number of voters who will vote for a particular political party and estimate which party will win
the election. Because the sample's finding is directly applied to the entire voters, the researchers must use a true-
random sampling to select participants. The key is that the sample should be taken from the voters to whom the
researchers intend to generalize using true-random sampling, which is presumed to accurately represent the entire
Mulisa, F. 78

voters. Otherwise, the results of the survey may be skewed, resulting in inaccurate predictions about which
political party will win the election. Indeed, Adam (2020) stated that when a sample does not accurately represent
the population from which it was drawn, it is unlikely to achieve valid inferences about the population
characteristics. Similarly, if the findings of the study are to be applied directly to the population, true-random
sampling methods are recommended. The following are the most common approaches to true-random sampling:

Simple random sampling

This is a sampling method that requires the compilation of a list of all units of the population, despite the fact that
the population is often small. Researchers select the sample using a variety of techniques based on a complete
population list. In a simple random sampling, all units of the population have the same chance of being included
in a particular study (Berndt, 2020; Salganik & Heckathorn, 2004). The sample is selected accordingly by giving each
unit of the population an equal opportunity to be included in the sample. In particular, this method is believed to
be more accurate and should be considered better if the population is homogeneous (Singh & Masuku, 2014). The
fundamental assumption is that the desired features are spread evenly throughout the target population. As a
result, the results of the sample can readily be extrapolated to the entire population (Delice, 2010). This might also
be because researchers that rely on a quantitative approach frequently adopt this method (Sargeant, 2012).
Applying simple random sampling methods, the most common approaches are lottery methods and random
number methods (Elfil & Negida, 2017).

Unadjusted systematic sampling

If the population is heterogeneous and large in size, it is a useful method to use unadjusted systematic random
sampling methods. Unadjusted refers to approaches that strictly adhere to theoretical principles without
modification. In this sampling technique, the initial unit of the sample is chosen at random, and the subsequent
units are chosen systematically. The researchers choose the units that will be included in the sample based on a
predetermined interval, which is commonly regarded as a scientific procedure (Elfil & Negida, 2017). The most
significant advantage of this technique is that it is less time-consuming to implement than simple random sampling
and tends to choose more evenly across the population (Berndt, 2020). If the population has N units and n units
are to be chosen, the sample interval will be computed as N/n, which is commonly known as k th. The first unit is
chosen at random from the sampling interval, and the successive units are processed by selecting the k th unit from
each sample interval. Every kth is selected directly without replacement to represent the entire population in
unadjusted systematic sampling. However, if changes are made to the sample selection procedures and the
method's true randomness is violated, the sample's accuracy in directly representing the population will be
compromised. Instead, such sampling approaches can be regarded as quasi-sampling, which can be indirectly
generalized to the entire population.

Quasi-random sampling techniques

True-random sampling is mostly guided by the availability of complete sampling frames, which does not always
work because it can be difficult to obtain a complete sampling frame in some circumstances. As a result,
researchers make every effort to ensure a random selection of samples, even if they do not have a complete
sampling frame. On the one hand, there is no complete sampling frame, making it difficult to implement, and on
the other hand, the researchers make every effort to ensure that the selection is random. Such a technique for
sampling is characterized as quasi-random sampling. Nonetheless, once the sample frame is complete, a few of
the randomly selected units may refuse to participate in the study. As a result, researchers usually make certain
adjustments to random sample techniques, violating the true-random sampling assumption. To demonstrate how
widespread this practice is, Martnez-mesa et al. (2016) and Chander (2017) contend that researchers might need
to back up at least 10% of the sample size to remedy refusals or nonresponsive sample units. Furthermore, Israel
(2003) argues that it can be increased by 30% to compensate for non-response. Although such adjustments are

International Journal of Academic Research in Education, 8(1)

Sampling techniques involving human subjects … 79

common for researchers to substitute unresponsive participants and replace missing data, it appears that they are
not applying the principles of true random sampling. Rather, quasi-random is a type of sampling that is partially
random but fails to meet certain criteria of true randomization.

The argument behind the principle of quasi-random sampling is that it cannot be a true representative of the
population. The justification is that the sample was selected by different sampling methods that compromised the
application of true-random sampling. Therefore, such sampling methods may be more appropriate when the data
collected from the sample are indirectly generalized to the population rather than directly generalized to the
population. For example, research in social sciences such as business, education, and psychology is usually based
on surveys. While this may not always be the case, researchers in these disciplines frequently claim to use random
assignment of participants in their studies. The random assignment of participants to a particular study can yield
a valid result if the sample size is large enough. However, the concern is whether the findings can be applied
directly to the population from which the sample was drawn. In other words, using quasi-sampling is appropriate
when the theory of the findings is indirectly generalized to the population. This means that the findings of the
sample are theoretically generalized to the population rather than empirically. On the contrary, it necessitates the
use of true-random sampling to apply the findings of the sample to the entire population directly. It is also
important to determine whether the study's goal is to generalize its findings or advance theoretical understanding.
In some social science fields, such as psychology, there is a strong predisposition toward the development of
theories rather than the generalization of results to a specific population. In such a context, the quasi-random
selection of participants seems very conventional. The accounts of quasi-random sampling techniques will be
covered in the following section.

Adjusted systematic sampling

The technique of systematic random sampling is based on the selection of units in a specified gap known as the
sampling interval (Martnez-mesa et al., 2016). That means the term "systematic sampling" itself refers to the fact
that each chosen unit in the sample has a gap or interval among them. However, selecting each unit in the
population while strictly adhering to the interval may not always be possible, particularly if the sample frame is
incomplete, the population is large, cases are inaccessible, or nonresponding units exist (Berndt, 2020; Elfil &
Negida, 2017; Martnez-mesa et al., 2016). In such circumstances, researchers are required to replace
nonresponding or inaccessible participants. This kind of systematic sampling approach is regarded as adjusted as
researchers make changes to true-random. This method is similar to unadjusted systematic random sampling in
that the investigators select participants to be included in the sample based on a systematic rule and a specified
interval. However, for a variety of reasons, researchers adjust the selection of each unit by compromising the k th
cases.

Stratified random sampling

Stratified sampling divides the population into homogeneous, mutually exclusive subgroups known as strata (Elfil
& Negida, 2017). Before selecting a sample, a population can be segmented into a specified variable of interest
that has a value for all units in the sample frame (Taherdoost, 2016). In this type of sampling, all units of the
population have a known chance of being selected rather than an equal chance of being selected. If there is indeed
a random selection, the chance of being selected based on equal chance is limited to specific strata rather than
the entire population. Any of the sampling procedures can be used within each stratum. The sampling procedure
may differ from one stratum to the next. In this method of sampling, the strata are chosen using true-random
sampling rather than every unit of the population. As a result, this sampling technique is better known as quasi-
random sampling.
Mulisa, F. 80

Combined sampling

A combined sampling technique is a means of selecting samples using more than one sampling technique at a time.
Such sampling methods use a combination of random and nonrandom sampling procedures (Etikan & Bala, 2017).
Researchers who work on a mixed research approach frequently use this type of sampling technique. In the mixed
research approach, Palinkas et al. (2016) argue that whenever the study is large and complex, a single
methodological approach is frequently insufficient. As a result, to answer their research questions, researchers
employ multiple sampling techniques.

Cluster sampling

Cluster sampling is a method of dividing a population into distinct groups known as clusters (Singh & Masuku,
2014). Because it is impractical to incorporate all the clusters, some of the clusters are selected at random to
represent all the entire clusters. Then all units within the selected clusters are included in the sample (Taherdoost,
2016). There are no units from non-selected groups included in the sample, which means they are represented by
those from selected clusters. In this method, the clusters are randomly selected, but all units of the population do
not have an equal chance of being selected equally. As a consequence, it is unreasonable to regard this procedure
as true-random sampling.

Multistage sampling

Multistage sampling refers to the selection of samples that are grouped into various structures using progressively
smaller sampling units from each structure. Multistage sampling is a more complex version of cluster sampling and
requires structuring the participants into homogeneous subgroups (clusters). Multistage sampling is different from
cluster sampling in that it selects a sample from each cluster rather than including all units from the selected
clusters (Singh & Masuku, 2014). This type of sampling requires at least two stages. In the first round, large clusters
are spotted and framed. In the second stage, units are picked from the identified clusters using any of the
probability sampling strategies. Because all units in the overall structure are not allowed to equally engage in the
study, this method is better regarded as quasi-random.

Non-random sampling techniques

Unlike methods that use random sampling, in non-random sampling, participants are not chosen for a particular
study based on a random technique. Rather, it is used when researchers decide to intentionally include some
participants based on a subjective judgment of their suitability (Berndt, 2020; Gill, 2020; Martnez-mesa et al., 2016;
Taherdoost, 2016). In such sampling approaches, the possibility of selecting individuals who represent the target
population is negligible. Because participants are chosen arbitrarily, it is impossible to estimate the probability that
each unit will be included in a sample. There is also no guarantee that every unit has been allowed to be included
in the sample. Since this sampling approach rarely yields a representative sample and precluding sample bias is
impossible, the findings obtained are not often generalized to the target population (Kadam & Bhalerao, 2010).
The justification is that participants who are chosen using a non-random sampling technique are selected
intentionally rather than by chance. Unrepresentative samples, on the other hand, might be valuable for certain
study objectives and can help answer certain research questions as well as provide new hypotheses (Martinez-
mesa et al., 2016). In other words, random sampling yields more breadth, whereas non-random sampling yields
more depth (Etikan & Bala, 2017).

Although its findings cannot be generalized, researchers often use nonrandom sampling for a variety of reasons,
including convenience, gaining access to chosen participants, or because data gathering is more possible with
limited financial access (Etikan & Bala, 2017). Nonrandom sampling is used as an alternative to random sampling
when random sampling is not feasible (Ary et al., 2010). Furthermore, there are instances when researchers favor

International Journal of Academic Research in Education, 8(1)

Sampling techniques involving human subjects … 81

nonrandom sampling, predominantly when they do not intend to generalize the study's findings to the entire
population. Although studies that use non-random sampling are frequently useful (Curry et al., 2009; Gill, 2020),
the trustworthiness of such studies is determined by the participants' credibility, transferability, dependability, and
confirmability (Campbell et al., 2020). Therefore, it is indispensable to explain both the philosophical and
methodological motives for selecting particular participants in a particular study. The following are some of the
most widely used random sampling techniques in research involving human beings.

Convenience sampling

Convenience sampling is one of the nonprobability sampling techniques in which samples are chosen from the
population just because they are readily accessible to the researchers (Elfil & Negida, 2017). They frequently use
this technique because it is easy to select the sample, and the sample is not intended to be representative of the
entire population. It is widely used by qualitative researchers due to its time and cost-effectiveness, as well as the
convenience with which samples are accessible (Taherdoost, 2016).

Judgmental sampling

The phrases judgmental sampling and purposive sampling are treated the same way in this article. It represents a
type of non-probability sampling where researchers choose individuals from the population to take part in their
studies using their own discretion. If researchers know the composition and characteristics of the population, they
will most likely use judgmental or purposive sampling (Elfil & Negida, 2017). The justification is that selecting
participants from the population is reasonably decided by researchers who are familiar with the population
(Campbell et al., 2020). To put it differently, researchers intentionally choose people who are deemed competent
enough to answer research questions for a specific objective (Etikan & Bala, 2017). However, judgment sampling
is vulnerable to the researcher's biases and may be even more biased than convenience sampling. This is
particularly true if the preconception about the participants’ characteristics is incorrect. Regardless of this flaw,
judgmental sampling may be quite useful in exploratory research, such as selecting members for focus groups or
in-depth interviews. This form of sampling is crucial, especially if the population is diverse and including people
with opposing viewpoints is necessary.

Quota sampling

Quota sampling is a type of sampling technique in which researchers are compelled to include members of various
mutually exclusive subpopulations to meet the specified number of participants (Johnson & Christensen, 2020;
Singh & Masuku, 2014; Taherdoost, 2016). Quota sampling, like stratified sampling, is a technique used when the
study population is naturally divided into multiple distinct compositions. When a certain number of participants
(quotas) from the sub-populations is required, the researchers deliberately select the required number of
participants from each sub-population (Campbell et al., 2020). It is a method of ensuring that the sample size
required for each subgroup is met. Unlike stratified sampling, which attempts to guarantee a random selection of
participants from each stratum, this type of sampling selects participants depending on who is available and willing
within each subpopulation.

Snowball sampling

Snowball sampling is a technique in which research participants participate in the selection of new participants for
the proposed study. It is a type of sampling in which one study participant suggests another as a potential
participant in the study. In this method, researchers frequently ask each participant to provide them with access
to other participants who have the desired qualities (Elfil & Negida, 2017; Taherdoost, 2016). Snowball sampling
Mulisa, F. 82

is a good strategy to use if researchers are looking to find participants that are difficult to reach or who must meet
particular requirements (Berndt, 2020).

Table 1. A summary of true-random, quasi-random, and non-random sampling techniques

True-random Quasi-random Non-random

Every unit of the population's Equal Known Negligible
chance to be selected
Procedures for selecting Purely random Partly random Not random
samples for a study
Generalization of findings to the Direct Indirect Impossible
population
Some examples of sampling Simple random, Adjusted systematic Conventional
techniques unadjusted systematic sampling, stratified sampling,
sampling sampling, cluster judgmental
sampling, and multi- sampling, quota
stage sampling sampling, and
snowball sampling

Conclusions

The probability and nonprobability approaches are the two most widely used sampling techniques in research.
Probability sampling is based on the random selection of participants, whereas nonprobability sampling employs
a non-random sampling process. The classifications of sampling into random and non-random do not appear to be
sufficient to accommodate all the sampling procedures currently in use. For example, some authors assign research
participants to a study at random and claim to have used random sampling. While it may appear that randomly
assigning study participants may be necessary, it is not a sufficient condition for the technique to be random. To
compensate for such a shortcoming, categorizing sampling as true-random, quasi-random, or nonrandom appears
to be useful. It is indeed worth noting that the random sampling approach is separated into two types: true random
sampling and quasi-random sampling. True random selection is based on providing participants with an equal
chance of being selected and proceeding with the selection accordingly without compromising it. On the other
hand, quasi-random selection is predicated on a known chance of being chosen at the stratum or cluster level
rather than at the unit level. As is typical with non-random sampling, there is no possibility of being included in the
sample for the participants. Finally, this paper suggests that when the sample is a direct representation of the
population, true-random sampling is essential, whereas quasi-sampling is sufficient when the sample represents
the population indirectly through theory.

Conflict of interest
The authors have declared that no competing interests exist.
Funding
The author received no funding for this work.

International Journal of Academic Research in Education, 8(1)

Sampling techniques involving human subjects … 83

References

Adam, A. M. (2020). Sample size determination in survey research. Journal of Scientific Research & Reports, 26(5),
90–97. https://doi.org/10.9734/JSRR/2020/v26i530263
Ary, D., Jacobs, L. C., & Sorensen, C. (2010). Introduction to research in education (8th ed.). Wadsworth, Cengage
Learning.
Berndt, A. E. (2020). Sampling methods. Journal of Human Lactation, 36(2), 224–226.
https://doi.org/10.1177/0890334420906850
Bordens, K. S., & Abbott, B. B. (2018). Research design and methods: A process approach (10th ed.). McGraw-Hill
Education.
Campbell, S., Greenwood, M., Prior, S., Shearer, T., Walkem, K., Young, S., Bywaters, D., & Walker, K. (2020).
Purposive sampling: Complex or simple? Research case examples. Journal of Research in Nursing, 25(8),
652–661. https://doi.org/10.1177/1744987120927206
Cassady, R. (1945). Statistical sampling techniques and marketing research. Journal of Marketing, 9, 317–341.
https://doi.org/10.1177/002224294500900401
Chander, N. G. (2017). Sample size estimation. Journal of Indian Prosthodontic Society, 17(3), 217–218.
https://doi.org/10.4103/jips.jips_169_17
Cohen, L., Manion, L., & Morrison, K. (2018). Research methods in education (8th ed.). Routledge, Taylor and
Francis Group.
Curry, L. A., Nembhard, I. M., & Bradley, E. H. (2009). Qualitative and mixed methods provide unique contributions
to outcomes research. Circulation, 119, 1442–1452.
https://doi.org/10.1161/CIRCULATIONAHA.107.742775
Delice, A. (2010). The sampling issues in quantitative research. Educational Sciences: Theory & Practice, 10(4),
2001–2018. https://files.eric.ed.gov/fulltext/EJ919871.pdf
Elfil, M., & Negida, A. (2017). Sampling methods in clinical research; An educational review. Emergency, 5(1), e52.
https://doi.org/10.22037/emergency.v5i1.15215
Etikan, I., & Bala, K. (2017). Combination of probability random sampling method with non probability random
sampling method (sampling versus sampling methods). Biometrics & Biostatistics International Journal, 5(6),
210–213. https://doi.org/10.15406/bbij.2017.05.00148
Gill, S. L. (2020). Qualitative sampling methods. Journal of Human Lactation, 36(4), 579–581.
https://doi.org/10.1177/0890334420949218
Israel, G. D. (2003). Determining sample size.
https://www.tarleton.edu/academicassessment/documents/samplesize.pdf
Johnson, R. B., & Christensen, L. (2020). Educational research: Quantitative, qualitative, and mixed approaches
(7th ed.). Sage Publications, Inc.
Kadam, P., & Bhalerao, S. (2010). Sample size calculation. International Journal of Ayurveda Research, 1(1), 55.
https://doi.org/10.4103/0974-7788.59946
Khaldi, K. (2017). Quantitative, qualitative or mixed research: Which research paradigm to use? Journal of
Educational and Social Research, 7(2), 15–24. https://doi.org/10.5901/jesr.2017.v7n2p15
Martinez-mesa, J., González-chica, D. A., Duquia, R. P., Bastos, J. L., & Bonamigo, R. R. (2016). Sampling: How to
select participants in my research study? An Bras Dermatol, 91(3), 326–330.
https://doi.org/10.1590/abd1806-4841.20165254
Mulisa, F. (2022). When does a researcher choose a quantitative, qualitative, or mixed research approach?
Interchange, 53(1), 113–131. https://doi.org/10.1007/s10780-021-09447-z
Palinkas, L. A., Horwitz, S. M., Green, C. A., Wisdom, J. P., Duan, N., Hoagwood, K., Angeles, L., & Northwest, K. P.
(2015). Purposeful sampling for qualitative data collection and analysis in mixed method implementation
research. Administration and Policy in Mental Health, 42(5), 533–544. https://doi.org/10.1007/s10488-013-
0528-y
Mulisa, F. 84

Salganik, M. J., & Heckathorn, D. D. (2004). Sampling and estimation in hidden populations using respondent-
driven sampling. Sociological Methodology, 34(1), 193–239. https://doi.org/10.1111/j.0081-
1750.2004.00152.x
Sargeant, J. (2012). Qualitative research part II: Participants, analysis, and quality assurance. Journal of Graduate
Medical Education, 4(1), 1–3. https://doi.org/10.4300/jgme-d-11-00307.1
Singh, A. S., & Masuku, M. B. (2014). Sampling techniques & determination of sample size in applied statistics
research: An overview. International Journal of Economics, Commerce and Management, 2(11), 1–22.
http://ijecm.co.uk/wp-content/uploads/2014/11/21131.pdf
Taherdoost, H. (2016). Sampling methods in research methodology; How to choose a sampling technique for
research. International Journal of Academic Research in Management (IJARM), 5(2), 18–27.
https://doi.org/10.2139/ssrn.3205035
Walters, W. H. (2021). Survey design, sampling, and significance testing: Key issues. Journal of Academic
Librarianship, 47(3), 1–9. https://doi.org/10.1016/j.acalib.2021.102344

International Journal of Academic Research in Education, 8(1)