Sampling Strategies for Quantitative and Qualitative Business Research

Sampling Strategies for Quantitative and Qualitative Business

Research
Vivien Lee, Psychology, University of Minnesota, and Richard N. Landers, Psychology, University
of Minnesota

https://doi.org/10.1093/acrefore/9780190224851.013.216
Published online: 23 March 2022

Summary
Sampling refers to the process used to identify and select cases for analysis (i.e., a sample) with the goal of drawing
meaningful research conclusions. Sampling is integral to the overall research process as it has substantial
implications on the quality of research findings. Inappropriate sampling techniques can lead to problems of
interpretation, such as drawing invalid conclusions about a population. Whereas sampling in quantitative research
focuses on maximizing the statistical representativeness of a population by a chosen sample, sampling in
qualitative research generally focuses on the complete representation of a phenomenon of interest. Because of this
core difference in purpose, many sampling considerations differ between qualitative and quantitative approaches
despite a shared general purpose: careful selection of cases to maximize the validity of conclusions.

Achieving generalizability, the extent to which observed effects from one study can be used to predict the same and
similar effects in different contexts, drives most quantitative research. Obtaining a representative sample with
characteristics that reflect a targeted population is critical to making accurate statistical inferences, which is core to
such research. Such samples can be best acquired through probability sampling, a procedure in which all members
of the target population have a known and random chance of being selected. However, probability sampling
techniques are uncommon in modern quantitative research because of practical constraints; non-probability
sampling, such as by convenience, is now normative. When sampling this way, special attention should be given to
statistical implications of issues such as range restriction and omitted variable bias. In either case, careful planning
is required to estimate an appropriate sample size before the start of data collection.

In contrast to generalizability, transferability, the degree to which study findings can be applied to other contexts, is
the goal of most qualitative research. This approach is more concerned with providing information to readers and
less concerned with making generalizable broad claims for readers. Similar to quantitative research, choosing a
population and sample are critical for qualitative research, to help readers determine likelihood of transfer, yet
representativeness is not as crucial. Sample size determination in qualitative research is drastically different from
that of quantitative research, because sample size determination should occur during data collection, in an ongoing
process in search of saturation, which focuses on achieving theoretical completeness instead of maximizing the
quality of statistical inference.

Theoretically speaking, although quantitative and qualitative research have distinct statistical underpinnings that
should drive different sampling requirements, in practice they both heavily rely on non-probability samples, and
the implications of non-probability sampling is often not well understood. Although non-probability samples do not
automatically generate poor-quality data, incomplete consideration of case selection strategy can harm the validity
of research conclusions. The nature and number of cases collected must be determined cautiously to respect

Page 1 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

research goals and the underlying scientific paradigm employed. Understanding the commonalities and differences
in sampling between quantitative and qualitative research can help researchers better identify high-quality
research designs across paradigms.

Keywords: sampling, non-probability sampling, convenience sampling, sample size, quantitative research, qualitative
research

Subjects: Research Methods

Sampling

Sampling is the process of identifying and selecting cases for data collection and analysis. It is a
necessary step in most research because it is neither realistic nor feasible to collect information
from every person or source of potential relevance to a research question given time, financial, and
personnel limitations (Adams et al., 2007). In general, sampling strategy should be integrated into
the logic of any research study, and the rationale for sample selection must align with the specific
goals of the study. Importantly, sampling is a multistep and premeditated process with tremendous
impact on the quality of data as well as the conclusions drawn from those data. Poor or
inappropriate sampling practices, such as choosing a nonrepresentative sample, can undermine
research findings (Short et al., 2002). Although there are some techniques a researcher can use to
remedy the effects of poor sampling practices (Katz & Vainas, 2021), a better approach is to avoid
poor sampling in the first place. A sampling strategy should be designed early in research planning,
alongside other validity concerns.

From a researcher’s perspective, sampling generally occurs over three main steps: (a) defining a
target population, (b) choosing a sampling frame, and (c) selecting samples from the sampling
frame (Bhattacherjee, 2012). First, a target population is identified. “Population” refers to the group
of people, events, observations, or phenomena one is interested in studying (Blackstone, 2012, p.
78). For example, this could be defined by the researcher as all employees at an organization, all
organizations within an industry, or all brands manufacturing a particular product. Second,
researchers identify a sampling frame, which is a subset of the target population from which
researchers can realistically collect data. In some cases, the sampling frame may encompass all
members of the target population; in others, the sample frame only contains a subset of the target
population, potentially leading to coverage error, which occurs when members of the sampling
frame do not randomly reflect those within the population of interest (Mazzocchi, 2008).
Identifying a representative sampling frame is especially critical when research goals involve
making inferences about target populations, which is usually the goal of quantitative research.
Third and finally, researchers must obtain a sample from the sampling frame, using a sampling
technique. Techniques can generally be classified as probability sampling, in which every person in
the sampling frame has a known probability to be selected, or non-probability sampling, in which
the probabilities of being selected are not known.

Page 2 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Sampling in Qualitative and Quantitative Research

The specific goals of sampling differ between quantitative and qualitative research approaches.
Quantitative research is used to develop and test theory describing precise, numerical relationships
among population parameters. This precision is generally believed by quantitative researchers to
produce results that are more objective, rigorous, and trustworthy. By contrast, qualitative research
is usually used to inductively develop theory describing complex and often highly contextualized
phenomena through the interpretation of rich source material (Polit & Beck, 2010). Sometimes, this
interpretation is used to generate quantitative hypotheses for later research or to explain and guide
the discovery of new phenomena (Mack et al., 2005). Less frequently, qualitative research is used
deductively, to test existing theories and hypotheses. This technique, called pattern matching,
requires researchers to observe the congruence of patterns of observations with those patterns
predicted by existing theory (Bitektine, 2008; Gilgun, 2013). Mixed methods research—a
combination of both quantitative and qualitative research in the same study—has also become
increasingly popular among business researchers in recent years. The ability of mixed methods
research to provide a deeper and broader range of information than either quantitative or qualitative
research alone has brought additive benefits in examining different questions (Almalki, 2016;
Creswell & Plano Clark, 2011). Ultimately, researchers should select a research approach that they
think best aligns with the specific research questions to be addressed.

Unique to quantitative research is its focus on generalizability, which demands the use of
representative samples that are ideally acquired through probability sampling. Generalizability can
be defined as “inferences about the conditions under which causal relationships hold across
persons, settings, treatment, and measurement variables” (Shadish et al., 2002). In other words,
generalizability can be conceptualized as the degree to which observed effects from one study can be
used to predict the same and similar effects in different contexts. The most straightforward way to
increase the probability of generalizable claims is to collect a representative sample that closely
mirrors the characteristics of the target population. To illustrate this kind of approach, consider a
study in which the goal is to generalize from a sample of volunteer study participants to a
theoretical population of people. If women in that population volunteered to participate at a greater
rate than men did, any statistics calculated from sample data would disproportionately reflect
women, creating a gender bias in the study’s results. Thus, the emphasis in quantitative research is
on generalizability and obtaining representative samples.

Unique to qualitative research is its focus on transferability, a related but distinct concept from
generalizability (Lincoln & Guba, 1985). Transferability can be defined as the extent to which the
findings of a particular study have applicability in other contexts or with other subjects.
Maximizing transferability requires researchers to design and then provide comprehensive
information about the nature of the population studied, sometimes called thick description, so that
future researchers can decide for themselves how conclusions should be applied to understand new
populations of interest. This is in stark contrast to generalizability in quantitative research, where
arguments for generalizability are generally left entirely to the author of the original research.
Given this difference, qualitative researchers are generally not as concerned with how well study
findings generalize across contexts, leaving this to the judgment of future researchers. Hence,

Page 3 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

qualitative researchers generally make sampling decisions based on how case characteristics and
counts affect their ability to address their research questions instead of by a priori targeting specific
populations to which they hope to generalize.

Given the divergent research objectives and theoretical expectations in sampling for both
quantitative and qualitative research, the remainder of this article explores sampling approaches
across quantitative and qualitative research, at the level of sampling technique selection as well as
sample size determination. Important themes mentioned throughout are summarized in Table 1. It
is argued here that quality sampling forms a vital and frequently under-considered foundation for
drawing valid conclusions in both quantitative and qualitative research contexts. First, the most
desirable sampling technique when trying to maximize generalizability, “Probability Sampling,” is
explored. Second, “Non-Probability Sampling” is examined in depth as it is arguably the most
common type of sampling used in both quantitative and qualitative research. Third, common types
of non-probability samples across quantitative and qualitative research are reviewed, followed by
statistical considerations associated with the use of non-probability samples unique to quantitative
research. Fourth, procedures for choosing an appropriate sample size in qualitative and quantitative
research are explored, respectively. Finally, conclusion and recommendations on sampling practices
are presented.

Table 1. Sampling Differences in Quantitative and Qualitative Research

Quantitative research Qualitative research

Purpose of sampling Generalizability Transferability

Sample Representativeness Drives Yes Not Usually

conclusion Validity

Ideal sampling approach Probability sampling Probability sampling or non-probability

sampling, depending on goals

Typical sampling approach Convenience sampling Snowball sampling (non-probability)

(non-probability)

Criteria for sample Size Effect size, power, and alpha Saturation, determined during data
determination level, all estimated a priori collection

Probability Sampling

In probability sampling, which is the ideal sampling technique to maximize generalizability, each
member in the population has a known and non-zero probability of being included through random
selection (Crano et al., 2015). Fundamental to probability sampling is simple random sampling, in
which every case in the target population has an equal and random chance of being selected.

Page 4 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Random sampling affords a certain degree of generalizability because when one draws a random
sample from a representative sampling frame, the mean of the sample should not be systematically
different from that of the target population. The use of random samples also makes it possible to
depict the relationship between a sample and the underlying population quantitatively because of
availability of confidence intervals, in which the population parameter is likely to fall (Tyrer &
Heyman, 2016). Although most research requires additional procedures that are far more complex,
simple random sampling forms the basis of most, if not all, probability sampling techniques.
Beyond simple random sampling, three additional broad techniques for probability sampling may
be used: systematic sampling, cluster sampling, and stratified sampling. To mimic the effects of
simple random sampling under certain assumptions, each of these techniques adds restrictions or
allowances to simple random sampling.

When the target population and available population for sampling are somewhat different, inclusion
probabilities for each member of a population can be different. When members are assigned varying
probabilities of being chosen, this is known as unequal probability sampling. Unequal probability
sampling can be used to “oversample” subgroups in a population if researchers believe that these
subgroups of members are more uncertain when equal probability sampling is used (Tillé &
Whilhelm, 2017). By oversampling, a researcher can slightly change the population to which their
research will best generalize. For example, consider a study where a multinational manufacturing
company with hundreds of factories is interested in knowing whether job satisfaction has an impact
on turnover intention among its factory workers. Because it will not be cost-efficient to survey all
workers across all factories, sampling is performed to select a subset of workers from some
factories. The inclusion probabilities for each worker are known. Whereas workers at a smaller
factory will have an equal chance of getting sampled under equal probability sampling as their
counterparts at a bigger factory, they can have a higher probability of getting sampled under
unequal probability sampling, which allows researchers to observe the unbiased population
relationship between the two variables, not just workers who work at specific factories.

Although probability sampling is the ideal option in most cases, it is often infeasible and
impractical, thereby leading researchers to use non-probability sampling techniques. Unless the
actual composition of the entire target population is known, making it possible to assign
probabilities to every individual within that population, a probability sample will not mimic a target
population exactly. Thus, probability sampling allows one to draw statistical inferences only under
very strict conditions that are often not feasible. Furthermore, even when probability sampling
techniques are applied, the resulting samples may still exhibit characteristics of non-probability
samples. For example, if cases drop out of a study for nonrandom reasons, sample statistics may
become biased due to underrepresentation of the characteristics of the dropped-out cases
(Mellenbergh, 2019).

Given the practical difficulties faced in pursuit of high-quality probability sampling, such
techniques are not widely used in practice, especially when the population of interest is “employees
in general” or “businesses in general,” for which random sampling is essentially impossible for the
investigation of most research questions. For this reason, the probability sampling techniques
covered above are not covered in as much depth as the non-probability sampling techniques that
follow. Thus, for more details on probability sampling, see Bhattacherjee (2012), Sharma (2017), and

Page 5 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Taherdoost (2016). Additionally, the differential coverage given to the non-probability sampling
techniques here does not suggest that they are gold-standard sampling techniques; instead, it only
reflects the higher occurrence of such techniques in modern business research.

Non-Probability Sampling

In contrast to probability sampling, inclusion probability for each case is not known and cases are
selected nonrandomly in non-probability sampling (Mellenbergh, 2019). Researchers typically
resort to non-probability sampling when finding that a representative sampling frame beforehand
is not feasible (Galloway, 2005). For example, if a researcher wants to study how cultural diversity in
companies influences job applicants’ perceptions of those companies, it is impossible to identify all
members of the “possible job applicant” population to form a representative sampling frame and
assign probabilities. There are two broad ways in which non-probability sampling can be conducted
to address this problem: self-selection and other selection. In self-selection, members of
populations that might be included elect to be included. For example, CEOs might choose if their
businesses are included in an industry-wide survey. In such cases, the resulting sample cases may
systematically differ from cases not self-selected into the study. The norm in non-probability
sampling, however, is other selection, in which cases are chosen by others, like the researcher, based
on some criterion such as convenience and subject matter expertise. Because of its convenience and
relative cost-efficiency, non-probability sampling is widely adopted across various disciplines in
quantitative and qualitative business research, despite the interpretive and validity challenges it can
introduce (Yang et al., 2006).

The term “non-probability sampling” is often used alongside “convenience sampling,” which is a
type of non-probability sampling in which researchers select members for samples as convenient
and accessible. For instance, consider a study in logistics and supply chain management where one
would like to study factors that could influence shipping times and goods distribution. Ideally, one
would have access to all organizations across the world that have business in this function, and
would be able to assign inclusion probabilities to each organization; however, this is not realistic. A
more likely practice would be for the researcher to conveniently sample organizations they readily
have access to and collect information on predictors such as shipment distance and destination,
which could potentially affect shipping times. Convenience samples are used like this by both
quantitative and qualitative researchers across business studies.

One of the most common non-probability techniques is purposive sampling. Purposive sampling,
also known as judgment sampling or expert sampling, is particularly prevalent among qualitative
researchers. Purposive sampling involves identification and selection of individuals who are
experienced and knowledgeable about a phenomenon of interest that is central to the research
question (Creswell & Plano Clark, 2011). Therefore, researchers typically adopt purposive sampling
whenever they want to collect data from a subgroup that fits a predetermined set of criteria to
obtain appropriate and useful information (Kelly, 2010). In purposive sampling, researchers
“purposively” seek samples that they think, in their own judgment or the judgment of their
research team, are appropriate and would add value to the study. For example, consider a cosmetic
company trying to identify how consumers, between 40 and 60 years old, who are not already their

Page 6 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

customers perceive their brand. Researchers engaging in purposive sampling might ask store
patrons to complete a questionnaire or answer a few questions about their age and previous buying
habits when they visit the store section. Later, they might look at these responses and pick
individuals to invite to target the specific sample composition they are interested in. Consequently,
purposive sampling is highly prone to researcher bias because of the many subjective judgments
and assumptions researchers must make about prospective participants during selection.
Nevertheless, purposive sampling is still regarded as an acceptable practice because when there are
clear and predetermined criteria that researchers can follow, researcher bias can be mitigated
(Sharma, 2017).

Quota sampling is another popular non-probability sampling technique. Quota sampling is

commonly utilized when one is interested in sampling specific subgroups and has a minimum
sample size requirement (Campbell et al., 2020; Mason, 2002). It builds on the concept of purposive
sampling in that it also relies on researchers’ subjective judgments. In quota sampling, each
member in the target population or sampling frame is divided into mutually exclusive subgroups
based on shared characteristics, such as age, gender, and nationality at an individual level; or
industry type and company size at an organizational level. Members within each subgroup are then
further selected purposively by researchers. Quota sampling is thus similar to stratified random
sampling, a probability sampling technique, in that they both also start with subgroups that contain
members with shared attributes; the distinction between the two is that quota sampling uses a
researcher’s subjective judgment to decide how each subgroup should be represented in the sample,
rather than through random selection, as would be used for stratified random sampling. For
example, consider a city interested in understanding average annual income across neighborhoods.
Participants might first be segmented into subgroups by postal code. With the goal of producing a
representative sample, researchers might then designate quotas for each subgroup and sample
residents based on their own judgment. Much like with purposive sampling, a mixture of researcher
beliefs, understanding of existing research literature, and the objective reality of the population
influence these inclusion decisions. Hence, although quota sampling is intended to acquire a
“representative” sample by including members with different characteristics, it may not be truly
representative of our target population because of the absence of random selection (Yang &
Banamah, 2014).

Due to their prevalence in business and management research, a non-exhaustive list of common
non-probability sampling techniques is reviewed below. More specifically, student samples,
organizational samples, snowball samples, and samples from panel services are closely examined. In
addition, this section identifies limitations associated with convenience sampling that researchers
should consider whenever statistical inferences are of interest.

Common Sources of Convenience Samples

Students are arguably the most common type of convenience sample in many people-focused areas
of business research, such as management and marketing, although they are more frequently used
in quantitative research than in qualitative research (Ashraf & Merunka, 2017; Hanel & Vione, 2016;
Peterson, 2001; Peterson & Merunka, 2014). College students are now typically recruited through

Page 7 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

flyers, social media, enrolled courses, and human subject pools in which students are given course
credit or financially compensated for participation. It is an inexpensive and easy way for academic
researchers to recruit participants. In business research, student samples are often used to make
claims that generalize far beyond the student population that they sample, such as through claims
about populations of adults and populations of employees. For example, the relationship between
perfectionism and burnout among students in a college setting might be used to draw conclusions
about worker perfectionism and career burnout. Indeed, the use of student samples may be
problematic as students tend to be more homogeneous in their responses when compared to the
general population of adults (Peterson, 2001). Student samples in such cases may not seem ideal
compared to sampling workers directly, but research on various organizational constructs has
found that published studies using student-recruited samples usually have equally representative
samples compared to non-student-recruited samples (Wheeler et al., 2014). Furthermore, results
from studies using student-recruited samples generally do not differ from non-student-recruited
samples aside from having smaller effect sizes (Demerouti & Rispens, 2014). Therefore, student
samples do not automatically equate to non-representative samples. Careful evaluation of
generalizability is required in every case.

Organizational samples, defined as members (e.g., people and organizations) sampled from within
one or more organizations, are another common convenience sample used in both quantitative and
qualitative research. Whenever organizational samples are used, the sampling unit can be either
individuals or organizations. Organizational sampling has intuitive appeal, as it would appear that
exploring a phenomenon (i.e., qualitative) or testing hypotheses (i.e., quantitative) at an
organizational setting increases applicability to other organizations. However, whenever
organizational sampling using employees as unit is involved, employees from very few, sometimes
even one, organizations are included. For instance, consider researchers who want to examine the
effects of a reward program on employee motivation. They may implement the program at one
convenient organization, measuring employee motivation before, during, and after the program,
and drawing conclusions about program effectiveness from differences in motivation scores. At the
surface level, using an organization sample like this seems much more appealing to researchers and
lay audiences than a student sample. However, in organizational contexts, higher-order constructs
can influence the entire sample, such as industry, organization size, organizational culture,
organizational climates, work group compositions, leadership effectiveness, and a myriad of other
such variables (Landers & Behrend, 2015). Another related example is when one is interested in
studying the relationship between companies’ financial statements and their stock prices. In this
case, the sampling unit is organizations instead of people. If the goal is to make broad claims about
all companies, then sampling only a handful of organizations is not sufficient because of the
higher-order constructs mentioned above. Caution is warranted when organizational samples are
in use, regardless of whether the sampling unit is employees or organizations. Without exploring
potential interactions between study variables and organizational characteristics peculiar to the
organization(s), an organizational sample is not necessarily any more generalizable than a student
sample and may even be misleading.

Page 8 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Snowball samples are commonly used in qualitative research, where generalizability is less
important than transferability. They are most typically used when target populations are generally
difficult to identify and access (i.e., have low visibility) (Biernacki & Waldorf, 1981), such as a
population comprising members with domain-specific knowledge and expertise or a population
with rare traits, experiences, or behavior. Snowball sampling most often starts with either a limited
number of previously identified members of the target population or with a general call in an online
or local community of population members. As researchers conduct interviews with members, they
ask members to “nominate” other members in their network that they think could make potential
contributions to the study, and then invite these individuals to participate in the study. For example,
suppose a researcher is interested in the experiences of people who have suffered from violent acts
by others while at work. In this case, there is no a priori identifiable population because such
experiences are neither common nor announced publicly. Although this would harm
generalizability in a quantitative study, as there would be no way to assess the degree to which the
victims of violence they interview were representative of such victims in general, the increased
diversity of the sample in terms of experiences adds richness that provides nuance and additional
meaningful context to understand transferability in a qualitative study.

Another increasingly common source of convenience sampling is crowdsourced panel research, an

approach now widely used by both academic and organizational researchers (Chandler et al., 2019).
Crowdsourced panels comprise large groups of participants who have volunteered to be available to
complete research studies over the Internet. Two popular services providing access to such panels
are Amazon Mechanical Turk (MTurk) and Prolific. Panel members are typically offered financial
compensation for study participation. Use of such services can be desirable for generalizability,
relative to other convenience sampling strategies, because they enable researchers to gain access to
populations beyond their institutions and across geographical locations, an affordance that, when
considered carefully, may be used to increase representation of a population of interest. Doing so
successfully enables researchers to draw more valid inferences about a larger unreachable
population. Although data from crowdsourcing may be compromised for many reasons, including
intentional or unintentional repeated participation, these services generally yield quality
quantitative data (Woo et al., 2015). Crowdsourced panels may even yield high-quality qualitative
data for some types of research questions, especially those concerning the experiences of
crowdsourcing workers (e.g., Brawley & Pury, 2016; Schmidt, 2015).

Statistical Considerations for Quantitative Convenience Samples

Convenience sampling does not necessarily undermine data quality or strip findings of validity
despite common criticisms to that effect. In the case of qualitative research, the principal goal of
such research is usually not to produce generalizable results, but to provide a thick description
about the phenomenon of interest, and in turn to suggest conditions of transferability (Lincoln &
Guba, 1985). Convenience itself is thus not problematic as long as core research questions can be
reasonably addressed with available cases. However, for quantitative research, the effects associated
with the use of convenience samples on the validity of study findings should be assessed carefully

Page 9 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

on a case-by-case basis. Whereas generalizability in quantitative research is typically first

considered as a statistical concern, focused on the representativeness of sample statistics to
population parameters and transferability in qualitative research is more holistic.

Both generalizability and transferability depend largely on the characteristics of the sample as well
as the context of the research (Landers & Behrend, 2015). Rather than blindly condemning the use
of convenience samples in either situation, which may be practically impossible to avoid in pursuit
of some research questions, researchers should examine each application domain and research
purpose individually. In qualitative research, researchers are generally trained to consider these
nuances explicitly. Yet in quantitative research, researchers often overlook the statistical effects
associated with the use of convenience samples and the extent to which they limit generalizability
of findings by relying heavily on rules of thumb (Landers & Behrend, 2015). For example, a sample
from a convenient organization should not be automatically considered superior to any other type
of convenience sample, as any one organization may have characteristics that make findings from it
ungeneralizable to a broad population of organizations. In considering such issues, researchers
should explicitly consider two specific statistical dimensions of the problem associated with the use
of convenience samples: range restriction and omitted variables.

Range Restriction in Convenience Samples

First, convenience samples may be range restricted, which is common in organizational research in
which people in convenient organizations were hired nonrandomly. Range restriction refers to
samples in which the full range of population values from a variable of interest are not observable in
the sample. For example, consider a convenient organization in which management researchers
want to study whether pre-employment testing tools are effective for predicting applicants’ future
job performance. This organization has extremely high standards for applicants, based on the
results of an interview and a psychometric test battery. If researchers are studying relationships
between interview performance and any other variables, their statistical tests will evidence direct
range restriction, which occurs when the range of the variable of interest from the sample is
reduced because an external constraint (e.g., cutoff on test score) is imposed directly on that
variable. Because people were hired based on their interview performance, employees with low
interview performance will not exist in the available sample, biasing any such statistical tests.
Compared to direct range restriction, indirect range restriction, which refers to restriction on
correlates of directly range-restricted variables, is more likely to be overlooked. In this case, if
interview performance was correlated with leadership potential, leadership potential would be
indirectly range restricted despite not being used directly in hiring, and thus any statistical tests
involving leadership potential would be biased.

The influence of range restriction on study findings can be substantial, harming the validity of
conclusions based on the results (Hunter et al., 2006; Sackett & Yang, 2000). Varying range
restriction between studies can also create inconsistencies in results among existing studies on the
same population effects (Vaci et al., 2014). The effects of different types of range restriction can be
corrected using techniques such as Thorndike’s adjustment formulas (Thorndike, 1949) or the
missing data approach (Pfaffel et al., 2016). Importantly, range restriction only creates bias when
restricted observed variables are related to study variables in the population. For example, student
Page 10 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

samples are typically directly restricted on numerous variables such as age, years of education, and
work experience; organizational samples are typically directly restricted by company selection
processes and indirectly restricted on factors like industry function and organizational culture
(Landers & Behrend, 2015). Yet if range-restricted variables are not of interest to the research
question, observed effects will still be unbiased. This is a major reason that such evaluation must be
done on a case-by-case basis.

Omitted Bias in Convenience Sampling

Another statistical effect related to the use of convenience samples is omitted variable bias
(Antonakis et al., 2010; Sackett et al., 2003). Ultimately, even if they do not claim so, researchers are
interested in theorizing about causal relationships, and accurate conclusions about causation
requires that variables are not only unrestricted in range but also that their interrelationships do
not have alternative explanations caused by the sampling technique employed. For instance,
researchers often choose not to collect data on variables that are not central to their theorizing, but
moderate or covary with focal relationships and variables in their models. This can be extremely
problematic as it may give rise to endogeneity, model misspecification, and eventually bias model
estimates (Wooldridge, 2010). Most studies have omitted variables, but their biasing effects only
occur when omitted variables are related to both the outcome and another predictor (Hill et al.,
2021). For example, consider a study in marketing exploring the effects of a new type of
advertisement technology on purchase intentions. For this study, researchers conveniently sample
college students and randomly assign them to experience the technology or not, finding a large
difference between the two. Consider further, however, if the reason they found this large
relationship was because the impact of technology novelty among college students is biased upward
(i.e., college student characteristics interacting with the advertising technology) in comparison to
the general population of consumers to which researchers hope to generalize. Thus, had they
sampled from the full population consisting of possible customers from all ages, the effect would
have been much smaller because of the omitted and unmeasured moderating effect of the positive
attitude toward technological novelty among college students not found in the broader population.
In other words, omitted variable bias may present a challenge to convenience samples because
convenience samples often have characteristics that may interact with other variables of interest.
By choosing a non-convenience sample, omitted variable bias may be avoided.

The problem of endogeneity caused by omitted variable bias can be commonly rectified in four
major ways (Hill et al., 2021): (a) by using a proxy, (b) by incorporating control or instrumental
variables, (c) by adding fixed effects, or (d) by randomly assigning subjects so that the predictor will
not display systematic variation with the residual. However, these solutions are not always feasible
in business and management research. A more realistic remedy is to explicitly consider whether
omission would meaningfully affect conclusions given the specific goals of the research (Landers &
Behrend, 2015). Omitted variables have different implications in different research contexts,
sometimes biasing effects upward, sometimes biasing them downward, and sometimes having no
or little effect. For instance, if the goal of a research study was to compare the relative variance

Page 11 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

accounted for by a variety of different antecedents that were not already strongly theoretically
established as causal, omitted variables are likely to be more problematic than in a simple test of
statistical significance of an effect between randomly assigned groups (Meade et al., 2009).

Determining Sample Size

Determining an appropriate sample size is essential for both quantitative and qualitative studies.
Whereas sample size drives the diversity of data in qualitative studies, a transferability concern, it
drives statistical power and effect size measurement precision in quantitative studies (Brewerton &
Millward, 2001). An adequately powered study allows researchers to correctly reject null hypotheses
when such hypotheses should be rejected. Although the relationship between sample size and
power is not strictly linear, increasing sample size generally boosts power (Baguley, 2004), and an
underpowered design reduces a researcher’s ability to detect a true effect (i.e., increased type II error
rate). Thus, determining a sufficient sample size that makes the best use of available resources is
crucial. There are also clearer distinctions between quantitative and qualitative research when
determining sample sizes, in part because the effects of inadequate sample size are so different
across the two contexts.

Quantitative Research

In quantitative research, the problem of estimating sample sizes can be largely divided into two
types—necessary sample sizes for hypothesis testing and necessary sample sizes for estimating
population parameters with a desirable level of precision (Binu et al., 2014). Research using
hypothesis testing generally tests whether some measured parameter differs across two or more
subpopulations within the broader sampling frame. For instance, researchers might compare store
sales before and after a loyalty program is implemented to see if the observed difference is
meaningful. By contrast, sample size requirements for estimating parameters are generally driven
by the desired width of confidence intervals surrounding estimated parameters. In this same
example, researchers might need to know the population effectiveness of the loyalty program
within +/–5% to justify using it in forecasting models related to business strategy.

The more common approach in business and management research is hypothesis testing, in which
desired sample size should be calculated and determined based on an a priori power analysis
(Abraham & Russell, 2008; Baguley, 2004; Bartlett et al., 2001), which requires researchers to specify
the confidence level, desired power level, and anticipated effect size before data collection occurs
(Cohen, 1992). Confidence level is defined based on the type I error rate set by the researcher (i.e.,
alpha or false positive rate), which corresponds to the probability of incorrectly rejecting a null
hypothesis. Normally, if researchers want to minimize false positives, they will use a more
conservative and smaller alpha level like 0.001 or 0.01 as opposed to the conventional rate of 0.05
(Kim, 2018). There are currently no strict guidelines or common consensus on optimal type I error
rates across contexts, so the value is chosen by researchers arbitrarily, although certain values have
become norms within specific fields of study. The second element needed to compute a suitable
sample size is desired power, which is the probability of correctly rejecting a false null hypothesis.

Page 12 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Though common practice is to seek 80% power or greater, it is again a case-by-case consideration,
taking into account the specific research questions and practical constraints like budget or data
collection time (for an overview, see Hallahan & Rosenthal, 1996). Finally, an expected effect size
must be specified in order to estimate an appropriate sample size. The choice of effect size is
typically driven by prior literature or pilot studies.

In contrast to inferential research, studies that target estimating population parameters with a
certain degree of precision are descriptive in nature and require a different approach to estimating
required sample size. In these cases, researchers typically establish a desired error rate, sometimes
called margin of error, and then determine the necessary sample size to achieve that margin of
error. Margin of error specifies the maximum distance that a sample statistic may deviate from a
population parameter. It is related to confidence intervals in that higher margins of error are
reflected in wider confidence intervals.

Qualitative Research

For many years, qualitative research has been criticized for lack of rigor because of its typically
much smaller sample sizes in relation to most quantitative studies, a problem amplified by the
absence of sample size justification in published research (Marshall et al., 2013). Sample size
determination can vary drastically depending on the specific research question as well as the
qualitative paradigm in which the research is being conducted (Boddy, 2016). Most qualitative
studies follow an interpretive paradigm in which the goal of research is to understand phenomena
through the subjective experiences and/or perceptions of researchers, rather than by derivation of
universal laws that can be generalized across contexts (Alharahsheh & Pius, 2020; Antwi & Hamza,
2015). The focal interest is typically in what is specific and unique about a given phenomenon
(Pizam & Mansfeld, 2009). Observation and interpretation thus form the basis of the interpretive
paradigm. Qualitative research following this tradition is generally only concerned with sample size
insofar as getting a broader range of experiences helps gain a more nuanced understanding of the
phenomenon of interest. Thus, a sample size as small as one case can be used as long as it is
thoughtfully and philosophically justified (Boddy, 2016). However, when a positivist paradigm in
which the goal is to offer a strong explanation or prediction regarding a certain phenomenon is
adopted, a larger sample size is reasonably expected (Pizam & Mansfeld, 2009). That said, sample
size determination in qualitative research in modern research is typically an iterative process that
can be viewed at both a theoretical and a practical level.

Theoretically speaking, qualitative researchers generally decide on a sample size based on

saturation. When sample size is not determined a priori, as is common in qualitative research,
achieving saturation is the criterion for discontinuing data collection, an approach that has become
the contemporary “gold standard” for determining sample size (Cleary et al., 2014; Saunders et al.,
2018). There are many forms of saturation: theoretical saturation, thematic saturation, and data
saturation (Saunders et al., 2018; Sebele-Mpofu, 2020). Rooted in a qualitative approach called
grounded theory, theoretical saturation refers to the point at which collecting more data about a
theoretical category reveals no new properties or theoretical insights about the emerging grounded
theory (Charmaz, 2014). There are two subtypes of thematic saturation: inductive and a prior

Page 13 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

thematic saturation. These two forms of saturation differ in that the former is also rooted in the
grounded theory approach, but the latter is not. Inductive thematic saturation focuses on
identifying new themes bottom-up, whereas a prior thematic saturation focuses on using collected
data to support or illustrate predetermined, top-down categories. Data saturation, which can be
defined as the degree to which new data repeat collected data, is another common type of saturation.
Data saturation is attained when collecting new information becomes redundant. Although there
are various forms of saturation a researcher can adopt, the choice of a targeted form of saturation is
contingent on the specific study design and research objectives.

Researchers also often blend two or more forms of saturation, making the conceptualization of
saturation difficult and ambiguous (Saunders et al., 2018; Sebele-Mpofu, 2020). Many published
qualitative studies have not defined and operationalized what they meant by “saturation” in their
study; yet researchers should always clearly define and operationalize saturation in relation to their
research goals and design to justify their approach to sample size selection. Even with clear
operationalization, very few guidelines surrounding how to detect and establish saturation exist in
practice, making sample size determination a challenging task. The idea of using saturation to
choose a sample size for a study is undoubtedly helpful at a conceptual level, but the idea alone does
not provide any pragmatic guidelines as to when saturation is attained and thus an appropriate
sample size has been achieved (Guest et al., 2006).

To date, very few studies have examined how to determine whether saturation has been achieved in
practice. Brod et al. (2009) recommended tackling this problem by constructing a saturation grid,
where topics and themes (or categories) are used as columns and each row represents a new
interview group (or individual). Thematic saturation is considered established when the grid
column for the current group is empty, which suggests no new concepts have emerged. More
recently, Guest et al. (2020) developed a novel approach to evaluate thematic saturation. They
proposed a saturation ratio that can be calculated based on two elements: base size and run length.
When the saturation ratio is below or reaches the new information threshold, which is determined a
priori by researchers and indicates the percentage of new information, researchers can conclude
saturation has been achieved. Specifically, base size refers to the minimum number of interviews a
researcher should analyze to calculate the amount of gained information, whereas run length is the
number of interviews in which one looks for new information. The saturation ratio can then be
computed by dividing the number of new themes emerged during the base size interviews by the
number of new themes emerged in the run length interviews. A small saturation ratio implies not
much new information is accumulated from having more interviews, and hence it will make sense
to cease data collection. Although some guidelines, like those from Brod et al. (2009) and Guest et al.
(2020), are available, they are not common. There are no broadly accepted approaches, leaving the
ultimate approach to achieving saturation up to the researcher.

In sum, determining sample sizes in qualitative research using saturation is challenging. There is
no systematic and one-size-fits-all answer to what sample size is sufficient. It is up to qualitative
researchers to explain and defend their operationalization of saturation and explain how it is was
reached in relation to the study design and goals.

Page 14 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Conclusion

Sampling is a multistep process that deserves more meticulous consideration from researchers
during planning. Given the prevalence of both quantitative and qualitative research approaches in
business research, similarities and differences shared by the two research approaches should be
considered at different stages of the sampling process. In particular, researchers and/or readers
should embrace the reality that non-probability sampling techniques are nearly inevitable in
modern research because of various practical constraints, such as limited budget and labor, as stated
above. Thus, the use of non-probability samples in both quantitative and qualitative studies should
not be automatically condemned. Readers of qualitative researchers must understand that it is their
responsibility, and not that of the original researchers, to judge whether specific study findings
should be applied to a new setting. Similarly, readers of quantitative research should contemplate
the potential statistical implications (e.g., range restriction and omitted variables) of a specific
sample in relation to the research question at hand and use that to justify sampling strategy, as the
responsibility for justifying that conclusions from quantitative research are generalizable falls on
the researcher conducting the study.

There are several best practices that paper authors can use to make the sampling process more
transparent. In quantitative research reports, researchers should explore possible probability
samples before committing to non-probability sampling. When non-probability sampling is
unavoidable, they should carefully consider the statistical implications of the use of the specific
samples that are available (e.g., range restriction and omitted variables) and explicitly justify, when
describing their sample, why the validity of inferences drawn from such a sample should result in
generalizable conclusions. Therefore, sampling is an integral component to the process of research
planning. Moreover, quantitative researchers should also clearly state the rationale of using a
specific sample size in relation to the desired power, expected effect size, and confidence level. In
qualitative research reports, researchers should remind readers that assessing meaningful transfer
is the reader’s responsibility. The researcher’s role is to present suggestions for potential boundary
conditions of transfer as a decision aid to the reader. Closely related, qualitative researchers should
not make unjustifiably broad and generalized claims beyond what the sample collected can
reasonably support. Additionally, as stated in the APA “Journal Article Reporting Standards” (APA,
2020, pp. 2–4), a valuable resource for fully documenting sampling decisions, authors conducting
qualitative research should clearly state how they determined their specific sample size and, when
saturation is involved, how they operationalized and achieved it.

In general, researchers should always make the distinction between transferability goals in
qualitative research and generalizability goals in quantitative research. Just because the sampling
requirements for qualitative research are seemingly less stringent than those for quantitative
research does not make qualitative research inferior and less trustworthy. Rather, it highlights a
central distinction between the two approaches. Overall, sampling in quantitative and qualitative
research is a complex matter that has substantial implications for the validity of research findings
and deserves focused attention and discussion among business and management researchers.

Page 15 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Further Reading
Cornesse, C., Blom, A. G., Dutwin, D., Krosnick, J. A., de Leeuw, E., Legleye, S., Pasek, J., Pennay, D., Phillips, B.,
Sakshaug, J. W., Struminskaya, B., & Wenz, A. (2020). A review of conceptual approaches and empirical evidence on
probability and nonprobability sample survey research. Journal of Survey Statistics and Methodology, 8, 4–36.

Eriksson, P., & Kovalainen, A. (2008). Qualitative methods in business research <https://dx.doi.org/
10.4135/9780857028044>. SAGE.

Levitt, H. M. (2021). Qualitative generalization, not to the population but to the phenomenon: Reconceptualizing
variation in qualitative research <https://doi.org/10.1037/qup0000184>. Qualitative Psychology, 8(1), 95–110.

Nuttavuthisit, K. (2019). Qualitative consumer and marketing research: The Asian perspectives and practices. Springer.

Rahi, S. (2017). Research design and methods: A systematic review of research paradigms, sampling issues and
instruments development. International Journal of Economics and Management Sciences, 6(2), 1–5.

Walter, S. L., Seibert, S. E., Goering, D., & O’Boyle, E. H., Jr. (2019). A tale of two sample sources: Do results from online
panel data and conventional data converge? <https://doi.org/10.1007/s10869-018-9552-y> Journal of Business and
Psychology, 34(4), 425–452.

Wu C., & Thompson, M. E. (2020). Sampling theory and practice. Springer.

References
Abraham, W. T., & Russell, D. W. (2008). Statistical power analysis in psychological research <https://doi.org/10.1111/j.
1751-9004.2007.00052.x>. Social and Personality Psychology Compass, 2(1), 283–301.

Adams, J., Khan, H. T. A., Raeside, R., & White, D. (2007). Research methods for graduate business and social science
students. SAGE.

Alharahsheh, H. H., & Pius, A. (2020). A review of key paradigms: Positivism vs. interpretivism. Global Academic Journal
of Humanities and Social Science, 2(3), 39–43.

Almalki, S. (2016). Integrating quantitative and qualitative data in mixed methods research—Challenges and benefits.
Journal of Education and Learning, 5(3), 288–296.

American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.).

Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A review and
recommendations <https://doi.org/10.1016/j.leaqua.2010.10.010>. Leadership Quarterly, 21(6), 1086–1120.

Antwi, S. K., & Hamza, K. (2015). Qualitative and quantitative research paradigms in business research: A philosophical
reflection. European Journal of Business and Management, 7(3), 217–225.

Ashraf, R., & Merunka, D. (2017). The use and misuse of student samples: An empirical investigation of European
marketing research. Journal of Consumer Behaviour, 16(4), 295–308.

Page 16 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Baguley, T. (2004). Understanding statistical power in the context of applied research. Applied Ergonomics, 35(2), 73–
80.

Bartlett, J. E., Kotrlik, J. W., & Higgins, C. C. (2001). Organizational research: Determining appropriate sample size in
survey research. Information technology, Learning, and Performance Journal, 19(1), 43–50.

Bhattacherjee, A. (2012). Social science research: Principles, methods, and practices. Global Text Project.

Biernacki, P., & Waldorf, D. (1981). Snowball sampling: Problems and techniques of chain referral sampling <https://
doi.org/10.1177/004912418101000205>. Sociological Methods & Research, 10(2), 141–163.

Binu, V. S., Mayya, S. S., & Dhar, M. (2014). Some basic aspects of statistical methods and sample size determination in
health science research. Ayu, 35(2), 119–123.

Bitektine, A. (2008). Prospective case study design: Qualitative method for deductive theory testing <https://doi.org/
10.1177/1094428106292900>. Organizational Research Methods, 11(1), 160–180.

Blackstone, A. (2012). Principles of sociological inquiry—Qualitative and quantitative methods <https://

saylordotorg.github.io/text_principles-of-sociological-inquiry-qualitative-and-quantitative-methods/>. Saylor
Foundation.

Boddy, R. C. (2016). Sample size for qualitative research. Qualitative Market Research, 19(4), 426–432.

Brawley, A. M., & Pury, C. L. (2016). Work experiences on MTurk: Job satisfaction, turnover, and information sharing.
Computers in Human Behavior, 54, 531–546.

Brewerton, P. M., & Millward, L. J. (2001). Organizational research methods: A guide for students and researchers (1st
ed.). SAGE.

Brod, M., Tesler, L. E., & Christensen, T. L. (2009). Qualitative research and content validity: Developing best practices
based on science and experience. Quality of Life Research, 18(9), 1263–1278.

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 <https://doi.org/10.1177/1744987120927206>. Journal of
Research in Nursing, 25(8), 652–661.

Chandler, J, Rosenzweig, C., Moss, A. J., Robinson, J., & Litman, L. (2019). Online panels in social science research:
Expanding sampling methods beyond Mechanical Turk <https://doi.org/10.3758/s13428-019-01273-7>. Behavioral
Research Methods, 51(5), 2022–2038.

Charmaz, K. (2014). Constructing grounded theory (2nd ed.). SAGE.

Cleary, M., Horsfall, J., & Hayter, M. (2014). Data collection and sampling in qualitative research: Does size matter?
Journal of Advanced Nursing, 70(3), 473–475.

Cohen, J. (1992). Statistical power analysis. Current Directions in Psychological Sciences, 1(3), 98–100.

Crano, W. D., Brewer, M. B., & Lac, A. (2015). Principles and methods of social research (3rd ed.). Routledge.

Creswell, J. W., & Plano Clark, V. L. (2011). Designing and conducting mixed methods research (2nd ed.). SAGE.
Page 17 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Demerouti, E., & Rispens, S. (2014). Improving the image of student‐recruited samples: A commentary <https://
doi.org/10.1111/joop.12048>. Journal of Occupational and Organizational Psychology, 87(1), 34–41.

Galloway, A. (2005). Non-probability sampling. In K. Kempf-Leonard (Ed.), Encyclopedia of social measurement (pp.
859–864). Elsevier.

Gilgun, J. F. (2013). Grounded theory, deductive qualitative analysis, and social work research and practice <http://
www.jstor.org/stable/10.7312/fort16138.8>. In A. E. Fortune, W. J. Reid, & R. L. Miller (Eds.), Qualitative research in social
work (2nd ed., pp. 107–135). Columbia University Press.

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and
variability <https://doi.org/10.1177/1525822X05279903>. Field Methods, 18(1), 59–82.

Guest, G., Namey, E., & Chen, M. (2020). A simple method to assess and report thematic saturation in qualitative
research <https://doi.org/10.1371/journal.pone.0232076>. PLoS One, 15(5), e0230276.

Hallahan, M., & Rosenthal, R. (1996). Statistical power: Concepts, procedures, and applications <https://doi.org/
10.1016/0005-7967(95)00082-8>. Behaviour Research and Therapy, 34(5–6), 489–499.

Hanel, P. H. P., & Vione, K. C. (2016). Do student samples provide an accurate estimate of the general public? <https://
doi.org/10.1371/journal.pone.0168354> PLoS One, 11(12), e0168354.

Hill, A. D., Johnson, S. G., Greco, L. M., O’Boyle, E. H., & Walter, S. L. (2021). Endogeneity: A review and agenda for the
methodology-practice divide affecting micro and macro research <https://doi.org/10.1177/0149206320960533>.
Journal of Management, 47(1), 105–143.

Hunter, J. E., Schmidt, F. L., & Le, H. (2006). Implications of direct and indirect range restriction for meta-analysis
methods and findings <https://doi.org/10.1037/0021-9010.91.3.594>. Journal of Applied Psychology, 91(3), 594–612.

Katz, Y., & Vainas, O. (2021). Addressing non-representative surveys using multiple instance learning. In F. Zhu, B. X.
Ooi, & C. Miao (Eds.), Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (pp. 3117–
3127). Association for Computing Machinery.

Kelly, S. E. (2010). Qualitative interviewing techniques and styles. In I. Bourgeault, R. Dingwall, & R. de Vries (Ed.), The
Sage handbook of qualitative methods in health research (pp. 307–326). SAGE.

Kim, J. H. (2018). Tackling false positives in business research: A statistical toolbox with applications. Journal of
Economic Surveys, 33(3), 862–895.

Landers, R. N., & Behrend, T. S. (2015). An inconvenient truth: Arbitrary distinctions between organizational,
Mechanical Turk, and other convenience samples <https://doi.org/10.1017/iop.2015.13>. Industrial and Organizational
Psychology: Perspectives on Science and Practice, 8(2), 142–164.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. SAGE.

Mack, N., Woodsong, C., MacQueen, K. M., Guest, G., & Namey, E. (2005). Quantitative research methods: A data
collector’s field guide. Family Health International.

Page 18 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Marshall, B., Cardon, P., Poddar, A., & Fontenot, R. (2013). Does sample size matter in qualitative research? A review of
qualitative interviews in IS research. Journal of Computer Information Systems, 54(1), 11–22.

Mason, J. (2002). Qualitative researching (2nd ed.). SAGE.

Mazzocchi, M. (2008). Statistics for marketing and consumer research <https://www.doi.org/10.4135/9780857024657>.

SAGE.

Meade, A. W., Behrend, T. S., & Lance, C. E. (2009). Dr. StrangeLOVE, or: How I learned to stop worrying and love
omitted variables. In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and urban legends:
Doctrine, verity and fable in the organizational and social sciences (pp. 89–106). Taylor & Francis Group.

Mellenbergh, G. J. (2019). Nonprobability sampling <https://doi.org/10.1007/978-3-030-12272-0_3>. In G. J.

Mellenbergh (Ed.), Counteracting methodological errors in behavioral research (pp. 31–37). Springer.

Peterson, R. A. (2001). On the use of college students in social science research: Insights from a second-order meta-
analysis. Journal of Consumer Research, 28(3), 450–461.

Peterson, R. A., & Merunka, D. R. (2014). Convenience samples of college students and research
reproducibility <https://doi.org/10.1016/j.jbusres.2013.08.010>. Journal of Business Research, 67(5), 1035–1041.

Pfaffel, A., Kollmayer, M., Schober, B., & Spiel, C. (2016). A missing data approach to correct for direct and indirect
range restrictions with a dichotomous criterion: A simulation study. PLoS One, 11(3), e0152230.

Pizam, A., & Mansfeld, Y. (2009). Consumer behaviour in travel and tourism. Taylor and Francis Group.

Polit, D. F., & Beck, C. T. (2010). Generalization in quantitative and qualitative research: Myths and strategies <https://
doi.org/10.1016/j.ijnurstu.2010.06.004>. International Journal of Nursing Studies, 47(11), 1451–1458.

Sackett, P. R., Laczo, R. M., & Lippe, Z. P. (2003). Differential prediction and the use of multiple predictors: The omitted
variables problem <https://doi.org/10.1037/0021-9010.88.6.1046>. Journal of Applied Psychology, 88(6), 1046–1056.

Sackett, P. R., & Yang, H. (2000). Correction for range restriction: An expanded typology <https://doi.org/
10.1037/0021-9010.85.1.112>. Journal of Applied Psychology, 85(1), 112–118.

Saunders, B., Sim, J., Kingston, T., Baker, S., Waterfield, J., Bartlam, B., Burroughs, H., & Jinks, C. (2018). Saturation in
qualitative research: Exploring its conceptualization and operationalization <https://doi.org/10.1007/
s11135-017-0574-8>. Quality & Quantity, 52, 1893–1907.

Schmidt, G. B. (2015). Fifty days an MTurk worker: The social and motivational context for Amazon Mechanical Turk
workers. Industrial and Organizational Psychology, 8(2), 165–171.

Sebele-Mpofu, F. Y. (2020). Saturation controversy in qualitative research: Complexities and underlying assumptions. A
literature review. Cogent Social Sciences, 6(1), 1838706.

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized
causal inference. Houghton, Mifflin and Company.

Page 19 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Sharma, G. (2017). Pros and cons of different sampling techniques. International Journal of Applied Research, 3(7),
749–752.

Short, J. C., Ketchen, D. J., & Palmer, T. B. (2002). The role of sampling in strategic management research on
performance: A two-study analysis <https://doi.org/10.1177/014920630202800306>. Journal of Management, 28(3),
363–385.

Taherdoost, H. (2016). Sampling methods in research methodology; How to choose a sampling technique for research.
International Journal of Academic Research in Management, 5(2), 18–27.

Thorndike, R. L. (1949). Personnel selection; test and measurement techniques. Wiley.

Tillé, Y., & Wilhelm, M. (2017). Probability sampling designs: Principles for choice of design and balancing <http://
www.jstor.org/stable/26408224>. Statistical Science, 32(2), 176–189.

Tyrer, S., & Heyman, B. (2016). Sampling in epidemiological research: Issues, hazards and pitfalls <https://doi.org/
10.1192/pb.bp.114.050203>. BritJPsych Bulletin, 40(2), 57–60.

Vaci, N., Gula, B., & Bilalić. M. (2014). Restricting range restricts conclusions <https://doi.org/10.3389/fpsyg.
2014.00569>. Frontiers in Psychology, 5(569).

Wheeler, A. T., Shanine, K. K., Leon, M. R., & Whitman, M. V. (2014). Student-recruited samples in organizational
research: A review, analysis, and guidelines for future research. Journal of Occupational and Organizational
Psychology, 87(1), 1–26.

Woo, S., Keith, M., & Thornton, M. (2015). Amazon Mechanical Turk for industrial and organizational psychology:
Advantages, challenges, and practical recommendations <https://doi.org/10.1017/iop.2015.21>. Industrial and
Organizational Psychology, 8(2), 171–179.

Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT Press.

Yang, K., & Banamah, A. (2014). Quota sampling as an alternative to probability sampling? An experimental study.
Sociological Research Online, 19(1), 56–66.

Yang, Z., Wang, X., & Su, C. (2006). A review of research methodologies in international business. International Business
Review, 15(6), 601–617.

Related Articles
Missing Data in Research

Hypothesis Testing in Business Administration

Inferential Statistics

Effect Size and Effect Uncertainty in Organizational Research Methods

Control Variables in Management Research

Page 20 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025
Sampling Strategies for Quantitative and Qualitative Business Research

Page 21 of 21

Printed from Oxford Research Encyclopedias, Business and Management. Under the terms of the licence agreement, an individual
user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: University of Michigan; date: 28 March 2025