Qualitative Research Designs Sample Size
Qualitative Research Designs Sample Size
8(3)
DOI: 10.13187/jare.2021.3.60
Journal homepage:
http://kadint.net/our-journal.html
Articles
Abstract
Qualitative research is currently growing in acceptance, especially within the health research
scope. Notwithstanding this positive trend, issues about the adequacy of sample size have been a
contention among qualitative and quantitative-based researchers. Our paper seeks to address some
of the issues facing popular qualitative designs in human research with this backdrop. Our article
explains the five key qualitative designs (case study, narrative inquiry, ethnography,
phenomenology, and grounded theory). Based on the existing studies, we reported their respective
sample size ranges that supported their data adequacy points. Our paper posits that sample size
concerns for qualitative designs revolve around their extensiveness and appropriateness.
Therefore, qualitative researchers’ judgement for data adequacy for a particular method should not
only rely on data saturation or a rule-of-thumb. Instead, they should also be guided by their
research goals, sampling approach, and research participants. Furthermore, we recommend that
qualitative researchers always verify the quality of data saturation by conducting additional
interviews and be more open in reporting their selected methodologies.
Keywords: case study, ethnography, grounded theory, narrative inquiry, phenomenology,
qualitative studies, sample size, sampling.
1. Introduction
Sample size determination in qualitative research is a topic of concern (Marshall, 1996).
The debate revolves around acceptable sample size and how representative the selected samples
are (Kuzel, 1992). Other scholars have also argued whether researchers can decide on the number
of participants sampled in a given study a priori or not (Sim et al., 2018). The sample size of any
research work is crucial in satisfying the scientific quality and ethical standards (Francis et al.,
2010). For instance, over-sampling and under-sampling wastes research funds and participants’
time. However, the use of inadequate samples that are needed carries ethical and scientific
*Corresponding author
E-mail addresses: [email protected] (J.O. Sarfo), [email protected] (T.P. Debrah),
[email protected] (N.I. Gbordzoe), [email protected] (W.T. Afful),
[email protected] (P. Obeng)
60
Journal of Advocacy, Research and Education. 2021. 8(3)
challenges as participants may only reflect a limited worldview that may not necessarily reflect the
general population of interest. Such data produces non-generalisable results and also wastes
research funds and time. The burden of offering adequate sample sizes in research has been one of
the major criticisms against qualitative studies.
One of the most acceptable standards in qualitative research is to allow the data to reach data
saturation (Creswell, 2013; Creswell, Creswell, 2018). Data saturation is a data adequacy point
where no new information could be obtained from participants in qualitative research (Creswell,
2013). Researchers like Morse (1995, 2007) argue that data saturation is ‘theoretical’ as the idea
that ‘true saturation’ of information may only be assumed. Thus, the claim that theoretical data
saturation point has arrived during qualitative research is relative to space and time. Amid all these
arguments, qualitative researchers are required to address how many participants are enough to
reach data adequacy and possible theoretical saturation. Failure to reach data saturation could
influence the quality of research and the study’s trustworthiness (Fusch, Ness, 2015).
In practice, data saturation can be problematic given the many research designs. As evident
in qualitative studies, data saturation for one study is not enough for another (Fusch, Ness, 2015;
Marshall, Rossman, 2011). For instance, the point at which data saturation is attained in a
phenomenological design differs from a case study or an ethnographic study (Creswell, 2013;
Creswell, Creswell, 2018). Also, data saturation could be determined by the available number of
participants in a target population and not necessarily the amount of information. Arguably, as low
as six interviews may provide saturation in a population of six women with a particular lived
experience. Thus, data saturation is not about the number per se but the depth of data (Fusch,
Ness, 2015). Essentially, data saturation is about the quality of data and not the quantity of data,
although some rules of thumb for qualitative sample sizes suggest best practices for specific designs
(Creswell, 2013).
61
Journal of Advocacy, Research and Education. 2021. 8(3)
designs, sample sizes in case studies are typically small (Hammarberg et al., 2016; Vasileiou et al.,
2018). Sample size determination in qualitative case studies is mostly, if not always, inherent in
researchers’ definition of what a case is and the boundaries of the case (Gerring, 2004; Hyett et al.,
2014). Yin (2018) referred to this as ‘bounding the case’. By bounding the case to determine the
sample size, a researcher first clearly defines the specific unit of analysis. Available evidence on
qualitative case studies reveals that researchers who conduct case studies may decide to either
recruit a single unit of analysis or multiple units of analyses; this defines the sample size (Crowe et
al., 2011; Rashid et al., 2019; Schoch, 2016).
The sample size for a case study depends on the research question and the epistemological
assumption behind the research (Mills et al., 2009). For example, a single case approach may be
adequate when the researcher is interested in formulating an in-depth contextualisation and an
idiographic explanation of a research problem. As the name implies, a single unit case study mainly
focuses its interest on an individual case. Most researchers use it, operationalising a case study to a
sample of one. Contrary, Mills et al. (2009) proposed that when the researcher is interested in
using contrasting observations to provide more insight into propositions and replication of the
findings, a multiple case approach is recommended. In a multiple case study approach, cases
should be added until theoretical and information saturation. All the same, no specific rule dictates
the number of participants in a multiple case study (Paré, Elam, 1997). However, some literature
recommends using four to fifty cases for a multiple case study. Considering the focus of case study
design, too many cases in a multiple case study may lose the depth of information required (Mills
et al., 2009).
Narrative Inquiry
As applicable in most qualitative designs, there is no single or strict rule for determining the
appropriate sample size for a narrative inquiry (Francis et al., 2010; Vasileiou et al., 2018).
Primarily, narrative inquiry utilises purposive sampling techniques, where researchers focus on
obtaining detailed and rich information from ‘fit-for-purpose’ participants (Nigar, 2020; Palinkas,
2014). Since narrative inquiry seeks to learn more about the narrator’s culture, historical
experiences, identity, and lifestyle, the emphasis is not on large sample sizes. As such, many
narrative studies focus on one individual, and this individual is selected based on his or her ability
to provide an understanding of the issues being addressed in the survey (Haydon, van der Riet,
2014; Moen, 2006).
Based on the nature of information required, some researchers may consider the need to
involve more than one person in a narrative inquiry. Vygotsky (1978) argued that the researcher
goes beyond interviewing the isolated individual when the study aims to understand human
development and functioning. Specifically, a systematic review revealed three studies reporting
specific sample sizes between one and twenty-four in education studies and sample sizes ranging
from one to fifty-two at an average of two sites in health science (Guetterman, 2015).
Ethnography
Like other qualitative designs, a researcher using ethnography is not bound by the use of
specific sample sizes (Morgan-Trimmer, Moser, Korstjens, 2018; Wood, 2016). Most ethnographic
studies focus on offering a detailed description of a culture. Since the complete picture of the
definition of culture is complex, it may be impossible to derive almost all the essential information
from one person. This approach accounts for more than one participant recruitment, although one
participant may be used in some cases. Most importantly, when a researcher desires to obtain
information that is truly representative of a larger sample, the determination of overall sample size
in ethnographic studies is influenced by the culture sharing group (Jaimangal-Jones, 2014; Jones,
Smith, 2017). When the culture sharing group has an adequate number to be interviewed or may
constitute a size appropriate for focus group discussion, ethnographic researchers may include the
entire group within the sample (Davis, Johnson, 2008). The review of sample sizes with an
emphasis on qualitative designs by Guetterman (2015) expounds on specific sample size ranges
utilised in most ethnographic studies both within education and health sciences. Guetterman found
that ethnographic studies conducted in education had sample sizes ranging from six to thirty-three
with an average of twenty-three. The smallest sample was nineteen in health science, with 586
being the largest sample size.
62
Journal of Advocacy, Research and Education. 2021. 8(3)
Phenomenology
Phenomenological research describes the shared meaning of multiple people’s lived
experiences with a topic or phenomenon. According to Husserl (2012), phenomenology recognises
the underlying logic of human experiences and communicates that logic accurately. Typically,
phenomenological investigations begin with a query concerning a phenomenon. Hence,
a phenomenological study is an explanatory method that comprehensively explains what it means
to be human by studying the lived experiences of individuals or groups to build concepts (Creswell,
2013; Husserl, 2012; van Manen, 1990). The phenomenological researcher’s job is to “build” the
investigated object based on its manifestations, structures, and components (Creswell, 2013).
According to Morse (1995), participants in phenomenological research are interviewed by
researchers to provide a considerable amount of data. Consequently, fewer individuals, typically
between six and ten, are employed. Guetterman (2015) noted an average sample size of fifteen,
ranging between eight and thirty-one in educational research, while participants between eight and
fifty-two, with a mean sample size of twenty-five, were seen in health studies. Similarly, five to
twenty-five participants are suggested as adequate for phenomenological studies by Creswell
(2013). Although small sample sizes are common in phenomenological studies, it is essential to
emphasise that the researcher may need to keep adding individuals until saturation to fulfil the
study’s goal.
Grounded Theory
The ground theory aims to create a ‘theory’ about a phenomenon using interviews and other
data collecting methods such as observations, documents, and audiovisual material (Creswell,
2013; Guetterman, 2015). According to Strauss and Corbin (1998), a critical concept in grounded
theory is that it is indicative, and theories are developed based on evidence from people who have
gone through the process. As a qualitative research design, the grounded theory study attempts to
uncover a theory rather than focusing on individual stories and shared experiences. According to
qualitative research academics, the issue of “how many” in determining the sample size is one
complex topic to answer directly. It is a crucial aspect in assessing the quality and validity of
qualitative research (Spencer et al., 2004). Generally, data saturation as a principle guides data
adequacy in grounded theory design. This principle usually occurs after the theory is formulated.
Determining saturation points across most qualitative studies is fluid. Some authors have
proposed that data saturation for grounded studies could be reached at nine interviews. In
contrast, others suggest sixteen and twenty-four interviews (Aldiabat, Navenec, 2018). Specifically,
Creswell (2013) recommended twenty and thirty informants for a grounded theory study, while
Morse (2000) indicated twenty to thirty participants with two or three unstructured interviews per
person. Similarly, Marshall et al. (2013) recommended that the sample size for grounded theory
should generally include between twenty to thirty interviews following an analysis of eighty-three
studies. On the other hand, Thomson (2010) suggested that the average sample size for grounded
theory is twenty-five. Nevertheless, thirty interviews allow the researcher to ultimately construct
patterns, ideas, categories, characteristics, saturation, and the dimension of a particular
phenomenon. According to Thomson (2010), while saturation is expected to occur around the
tenth interview, it is good to verify saturation by doing more interviews.
6. Funding
None.
References
Aldiabat, Navenec, 2018 – Aldiabat, K.M., Navenec, C.L. (2018). Data saturation:
The mysterious step in grounded theory methodology. The Qualitative Report. 23(1): 245-261.
Bernard, 2000 – Bernard, H.R. (2000). Research methods in anthropology: Qualitative and
quantitative approaches (3rd ed.). Thousand Oaks, CA: Sage.
Boddy, 2005 – Boddy, C.R. (2005). Sample size for qualitative research. Qualitative Market
Research. 19: 426-432.
Corbin, Strauss, 2015 – Corbin, J., Strauss, A. (2015). Basics of qualitative research:
Techniques and principles for developing grounded theory. Thousand Oaks, CA: Sage.
Creswell, 2013 – Creswell, J.W. (2013). Qualitative inquiry and research design: Choosing
among five approaches (3rd ed.). Thousand Oaks, CA: Sage.
Creswell, Creswell, 2018 – Creswell, J.W., Creswell, J.D. (2018). Research design: Qualitative,
quantitative, and mixed method approach (5th ed). Los Angeles: SAGE Publication Inc.
Crowe et al., 2011 – Crowe, S., Cresswell, K., Robertson, A., Huby, G., Avery, A., Sheikh, A.
(2011). The case study approach. BMC Medical Research Methodology. 11(100): DOI:
doi.org/10.1186/1471-2288-11-100
Davis, Johnson, 2008 – Davis, W.R., Johnson, B.D. (2008). Prescription opioid use, misuse,
and diversion among street drug users in New York City. Drug and Alcohol Dependence. 92(1-3):
267-276.
Devers, Frankel, 2000 – Devers, K.J., Frankel, R.M. (2000). Study design in qualitative
research: Sampling and data collection strategies. Education for Health. 13(2): 263-271.
Francis et al., 2010 – Francis, J. J., Johnston, M., Robertson, C., Glidewell, L., Entwistle, V.,
Eccles, M.P., Grimshaw, J.M. (2010). What is an adequate sample size? Operationalising data
saturation for theory-based interview studies. Psychology and Health. 25(10): 1229-1245.
Fusch, Ness, 2015 – Fusch, P.I., Ness, L.R. (2015). Are we there yet? Data saturation in
qualitative research. The Qualitative Report. 20(9): 1408-1416.
Gerring, 2004 - Gerring, J. (2004). What is a case study and what is it good for?
The American Political Science Review. 98(2): 341-354.
Guetterman, 2015 - Guetterman, T. (2015). Descriptions of sampling practices within five
approaches to qualitative research in education and the health sciences. Forum Qualitative
Sozialforschung/Forum: Qualitative Social Research. 16(2): e25.
Hammarberg et al., 2016 – Hammarberg, K., Kirkman, M., de Lacey, S. (2016). Qualitative
research methods: when to use them and how to judge them. Human Reproduction. 31(3): 498-501.
Haydon, van der Riet, 2014 – Haydon, G., Riet, P.V.D. (2014). A narrative inquiry: How do
nurses respond to patients’ use of humour? Contemporary Nurse. 46(2): 197-205.
Husserl, 2012 – Husserl, E. (2012). Ideas: General introduction to pure phenomenology.
Routledge.
Hyett et al., 2014 – Hyett, N., Kenny, A., Dickson-Swift, V. (2014). Methodology or method?
A critical review of qualitative case study reports. International Journal of Qualitative Studies on
Health and Well-Being. 9(1): e23606.
Jaimangal-Jones, 2014 – Jaimangal-Jones, D. (2014). Utilising ethnography and participant
observation in festival and event research. International Journal of Event and Festival
Management. 5(1): 39-55.
Jones, Smith, 2017 – Jones, J., Smith, J. (2017). Ethnography: Challenges and opportunities.
Evidence-Based Nursing. 20(4): 98-100.
Kuzel, 1992 – Kuzel, A.J. (1992). Sampling in qualitative inquiry. In B. F. Crabtree and
W.L. Miller (Eds.). Doing qualitative research (pp. 31–44). Sage Publications, Inc.
Marshall et al., 2013 – 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.
Marshall, 1996 – Marshall, M.N. (1996). Sampling for qualitative research. Family Practice.
13(6): 522-526.
64
Journal of Advocacy, Research and Education. 2021. 8(3)
Marshall, Rossman, 2014 – Marshall, C., Rossman, G.B. (2014). Designing qualitative
research. Sage.
Mills et al, 2009 – Mills, A.J., Durepos, G., Wiebe, E. (Eds.). (2009). Encyclopedia of case
study research. Sage Publications.
Moen, 2006 – Moen, T. (2006). Reflections on the narrative research approach.
International Journal of Qualitative Methods. 5(4): 56-69.
Molchanova, 2019 — Molchanova, V.S. (2019). Modern slavery in India: The essence, forms,
distribution. Slavery: Theory and Practice. 4(1): 20-28.
Morgan-Trimmer, Wood, 2016 – Morgan-Trimmer, S., Wood, F. (2016). Ethnographic
methods for process evaluations of complex health behaviour interventions. Trials. 17(1): 1-11.
Morse, 1995 – Morse, J. (1995). The significance of saturation. Qualitative Health Research.
5: 147-149.
Morse, 2000 – Morse, J.M. (2000). Determining sample size. Qualitative Health Research.
10: 3-5.
Morse, 2007 – Morse, J.M. (2007). Quantitative influences on the presentation of qualitative
articles. Qualitative Health Research. 17: 147-148.
Moser, Korstjens, 2018 - Moser, A., Korstjens, I. (2018). Series: Practical guidance to
qualitative research. Part 3: Sampling, data collection and analysis. European Journal of General
Practice. 24(1): 9-18.
Nigar, 2020 – Nigar, N. (2020). Hermeneutic phenomenological narrative enquiry:
A qualitative study design. Theory and Practice in Language Studies. 10(1): 10-18.
Onwuegbuzie, Leech, 2007 – Onwuegbuzie, A.J., Leech, N.L. (2007). A call for qualitative
power analyses. Quality and Quantity. 41: 105-121.
Palinkas, 2014 – Palinkas, L.A. (2014). Qualitative and mixed methods in mental health
services and implementation research. Journal of Clinical Child & Adolescent Psychology. 43(6):
851-861.
Paré, Elam, 1997 – Paré, G., Elam, J.J. (1997). Using case study research to build theories of
IT implementation. In Information systems and qualitative research. Boston, MA: Springer.
Pp. 542-568.
Rashid et al., 2019 – Rashid, Y., Rashid, A., Warraich, M.A., Sabir, S.S., Waseem, A. (2019).
Case study method: A step-by-step guide for business researchers. International Journal of
Qualitative Methods. 18. DOI: doi.org/10.1177/1609406919862424
Ritchie et al., 2014 – Ritchie, J.C., Lewis, J., Elam, G., Tennant, R., Rahim, N. (2014).
Designing and selecting samples. London: Sage.
Schoch, 2016 – Schoch, K.W. (2016). Case study research. In G.J. Burkholder, K.A. Cox, L.M.
Crawford, and J.H. Hitchcock (Eds.), The scholar-practitioner’s guide to research design.
Thousand Oaks, California: SAGE Publications, Inc.
Sim et al., 2018 – Sim, J., Saunders, B., Waterfield, J., Kingstone, T. (2018). Can sample size
in qualitative research be determined a priori? International Journal of Social Research
Methodology. 21(5): 619-634.
Spencer et al., 2004 – Spencer, L., Ritchie, J., Lewis, J., Dillon, L. (2004). Quality in
qualitative evaluation: A framework for assessing research evidence. UK: National Center for
Social Research.
Thomson, 2010 – Thomson, S.B. (2010). Sample size and grounded theory. Journal of
Administration and Governance. 5(1): 45-52.
van Manen, 1990 – van Manen, M. (1990). Researching lived experience: Human science for
an action sensitive pedagogy. State University of New York Press, Albany.
Vasileiou et al., 2018 – Vasileiou, K., Barnett, J., Thorpe, S., Young, T. (2018). Characterising
and justifying sample size sufficiency in interview-based studies: systematic analysis of qualitative
health research over a 15-year period. BMC Medical Research Methodology. 18(1): DOI:
10.1186/S12874-018-0594-7
Vygotsky, 1978 – Vygotsky, L.S. (1978). Mind in society: The development of higher
psychological processes. Cambridge, MA: Harvard University Press.
Yin, 2018 – Yin, R.K. (2018). Case study research: Design and methods (6th ed.). Thousand
Oaks, CA: Sage.
65