Sample Size and Grounded Theory

Abstract

Interviews are one of the most frequently used method of data collection and grounded theory has emerged as one of the most commonly used methodological frameworks. Although interviews are widely accepted, there is little written on an appropriate sample size. To tackle this concern a content analysis of one hundred articles that utilized grounded theory and interviews as a data collection method was performed. The findings indicate the point of theoretical saturation can be affected by the scope of the research question, the sensitivity of the phenomena, and the ability of the researcher. However, the average sample size was twenty-five, but it is recommended to plan for thirty interviews to fully develop patterns, concepts, categories, properties, and dimensions of the given phenomena. By knowing an approximation of the required number of interviews researchers now have starting point which will assist in the design, execution and budgeting of a research project.

Full text

45
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
Sample Size and Grounded Theory
S. B. Thomson
1
Abstract
Interviews are one of the most frequently used method of data collection and grounded
theory has emerged as one of the most commonly used methodological frameworks.
Although interviews are widely accepted, there is little written on an appropriate sample
size. To tackle this concern a content analysis of one hundred articles that utilized
grounded theory and interviews as a data collection method was performed. The findings
indicate the point of theoretical saturation can be affected by the scope of the research
question, the sensitivity of the phenomena, and the ability of the researcher. However,
the average sample size was twenty-five, but it is recommended to plan for thirty
interviews to fully develop patterns, concepts, categories, properties, and dimensions of
the given phenomena. By knowing an approximation of the required number of
interviews researchers now have starting point which will assist in the design, execution
and budgeting of a research project.
Keywords: Grounded theory, sample size, interviews, theoretical saturation
1
Assistant Professor in the Department of Business at St. George’s University in Grenada, West Indies.
Email: stanleybruce.thomson@gmail.com

46
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
Introduction
Qualitative research has increased in popularity in the last two decades and is becoming
widely accepted across most disciplines (e.g. sociology, medicine, business & economics,
psychology, anthropology) (Huberman & Miles, 2002). The data for qualitative research is
collected through interviews, field notes, observations, videos, personal journals, memos,
or other varieties of pictorial or written material with interviews being the most common
data collection method (Creswell, 1998; Marshall & Rossman, 2006; Strauss & Corbin,
1998). With the increased interest in qualitative research there are questions regarding
methodological issues (Jones & Noble, 2007; LaRossa, 2005). ‘Grounded theory’2 is one
of the most commonly used qualitative methods (Creswell, 1998). LaRossa (2005, p. 838)
argued that grounded theory methods were “exhilarating” but “extremely challenging”
due a perceived difficulty in understanding the procedure. In particular, sample size and
validity are the most often queried aspects of qualitative research.
The aim of this article is to provide an outline of sample size and the reason for these
requirements for grounded theory. In order to provide some empirical guidance for
estimating an appropriate sample size, one hundred articles were reviewed that used
grounded theory as a methodological framework3. The articles utilized interviews as a
data collection method. The first section will discuss the factors of sample size which will
be followed by a consideration investigation of theoretical sampling and its application in
grounded theory research.
Factors of Sample Size
The key to qualitative research and, in particular, grounded theory is to generate enough
data so that the illuminate patterns, concepts, categories, properties, and dimensions of
the given phenomena can emerge (Glaser & Strauss, 1967; Strauss & Corbin, 1998).
Therefore, it is essential to obtain an appropriate sample size that will generate sufficient
data (Auerbach & Silverstein, 2003). For example, if the researcher is studying how cash
incentives affect job retention within organization X which has 200 employees. The
researcher will have to interview an adequate number of employees so that a clear
picture of the patterns, concepts, categories, properties, and dimensions regarding how
cash incentives affect their desire to remain with organization X will emerge.
What is the appropriate sample size? The question is answered by the concept of
‘theoretical saturation’ (Glaser & Strauss, 1967; Strauss & Corbin, 1998). Theoretical
saturation occurs in data collection when:
2 Grounded theory is a process by which a researcher generates a theory that is grounded in the data (Glaser
& Strauss, 1967; Strauss & Corbin, 1998) by using a coding procedure to illuminate patterns or “concepts
that are the building blocks of theory” (Strauss & Corbin, 1998, p. 13). The procedure allows for a systematic
analysis of the data and follows a given, repeatable procedure.
3 A search was performed using Proquest ABInform with the search parameter of ‘grounded theory’ in the
citation and abstract. The first one hundred full text articles that stated sample size were used.

47
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
“(a) no new or relevant data seem to emerge regarding a category,
(b) the category is well developed in terms of its properties and dimensions
demonstrating variation, and
(c) the relationships among categories are well established and validated”
(Strauss & Corbin, 1998, p. 212).
In other words, the researcher continues expanding the sample size until data collection
(e.g. interviews) supplies no new data (Douglas, 2003; Goulding, 2002; Locke, 2001).
Hence continuing with the example, the researcher would continue interviewing
employees of organization X until the data they are gathering from the interviews
becomes repetitive i.e. no new data emerges. This might take 10, 20, 30 or more
interviews. By ignoring theoretical saturation the researcher risks creating theory based
on inadequate development of patterns or themes and the result might be findings based
on the lack of reliability and/or validity (Jones & Noble, 2007).
In the case of interviews, there is no set number for when theoretical saturation occurs
(Glaser & Strauss, 1967; Strauss & Corbin, 1998). One of the aspects is that sample size
dependents on the research question (Morse, 2000; Sobal, 2001). A broader research
scope will require far more data and thus require more data collection, which in turn,
requires to more interviews, and may require alternative data sources. This means
considerable more work for the researcher. Thus, Strauss & Corbin recommend
narrowing the focus of the research question at the beginning or after three or four
interviews (1998). By using the first few interviews as guides to the essence of the
phenomena the researcher can narrow the focus and thus reduce the number of
interviews (Kwortnik, 2003, Strauss & Corbin, 1998). The data gathered from the initial
interviews can be reviewed for later research projects that might focus on areas that
were seen as peripheral for the initial research project.
Another aspect that might dictate sample size is the nature or sensitivity of the
phenomena that being studied (Morse, 2000, Sobal, 2001). Values and beliefs that an
individual holds are often considered deeply personal and thus research participants
might be reluctant to share his or her thoughts. An example of this might be an
individual’s spiritual beliefs and how those beliefs relate to the work environment or how
an individual’s aligns his or her ethical values to those within the organizational
environment. In order to gain sufficient data to derive a deep understanding of perceived
sensitive and controversial phenomena, more interviews might be required (Morse,
2000). This could mean increasing the sample size or increasing the number of interviews
with each participant in order to generate a more open and trusting interview
environment. In turn, this trust would encourage a participant to be more forthcoming.
Conversely, the less sensitive the nature of the project is the easier it will be for
participants to talk about it (e.g. strategic planning implementation, job satisfaction)
(2000).
The ability, experience or knowledge of the researcher will also affect sample size
(Morse, 2000). Researchers with more experience and strong interviewing skills will
require fewer participants as they can guide and encourage a participant to reveal data
(Morse, 2000; Strauss & Corbin, 1998). The skill level of the researcher, usually acquired
through experience, puts the participant at ease, creates a more conversational
atmosphere that generates an aura of trust (Strauss & Corbin, 1998). The researcher’s

48
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
knowledge of the given area might provide the researcher with insight that allows them
to bypass unnecessary data and formulate questions that guide the interview more
effectively. The researcher’s knowledge can come from two sources; a literature review,
which might be scant in the case of new phenomena, or personal experience. For
example, Jette, Grover, and Keck (2003) used their experience and knowledge of physical
therapy to interview nine physical therapists and occupational therapists regarding
decision making in the discharge process in an acute care facility. A researcher with no
background in physical therapy would have required many more interviews in order to
gain sufficient background data to understand the decision making process.
Prior experience and knowledge could also be a hindrance to the researcher in that they
might be influenced by their prior knowledge and thus miss or put aside valuable insights
from participants (Strauss & Corbin, 1998). The preconceived notions that a researcher
has must act as only a guide in the first few interviews. Strauss & Corbin (1998, p. 205)
point out that:
“Because these early concepts have not evolved from “real” data, if the researcher
carries them with him or her into the field, then they must be considered
provisional and discarded as data begin to come in. Nevertheless, early concepts
often provide a departure point from which to begin data collection…”
Thus the factor that determines sample size is based on the concept of theoretical
saturation. As shown above the point of saturation can be affected by the scope of the
research question, the nature or sensitivity of the phenomena, and the ability, experience
or knowledge of the researcher. The following section will discuss the use of theoretical
sampling as a further determinate of sample size.
Theoretical Sampling
The quality of data can affect the sample size hence theoretical sampling is
recommended when using grounded theory. The theoretical sampling procedure dictates
that the researcher chooses participants who have experienced or are experiencing the
phenomenon under study. By doing so the researcher has chosen ‘experts’ in the
phenomenon and thus able to provide the best data (Corbin & Strauss, 1998; Glaser &
Strauss, 1967). However, the process of selecting participants is also an evolving process
based on the arising patterns, categories and dimensions emerging from the data.
Researchers seek out participants that might be able to provide deeper understanding of
the emerging patterns, categories and dimensions. Thus, if the participants are
predetermined the researcher might find that an increase in samples size is required in
order to follow emerging themes.
“Theoretical sampling is cumulative” (Strauss & Corbin, 1998, p. 203). Each interview
provides the researcher a selection of data on which he or she can build. Iterative analysis
of the collected interviews carried out through the data collection process allows the
researcher to visualize the emerging patterns, categories and dimensions (Kwortnik,
2003; Strauss & Corbin, 1998). Iterative analysis is a process in which the researcher
moves back and forth through the data in order to find, compare, and verify the patterns,
concepts, categories, properties and dimensions of the phenomena (Kwortnik, 2003).
Having gained an insight of what the emerging patterns, categories and dimensions are
the ensuing interviews can be focused on filling out those patterns, categories and
dimensions to the point of saturation. Therefore, sampling is also aimed at the same end

49
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
result of iterative analysis and might necessitate interviewing participants who are more
versed in the phenomena thus providing higher quality data (Glaser & Strauss, 1967;
Strauss & Corbin, 1998). The flexibility of theoretical sampling allows the researcher to
follow directions indicated by the data; hence, a reduction in sample size is possible by
choosing appropriate participants (Glaser & Strauss, 1967; Strauss & Corbin, 1998).
By using theoretical sampling and targeting the most knowledgeable participants the
quality of the data gathered in each interview can be increased. “There is an inverse
relationship between the amount of usable data obtained from each participant and the
number of participants” (Morse, 2000, p. 4). In other words, the greater the amount of
usable data a researcher is able to gather from a single participant the fewer participants
that will be required (Morse, 2000). Thus, theoretical sampling provides a sample
selection that is more likely to highlight the patterns, concepts, categories, properties,
and dimensions of the given phenomenon (Glaser & Strauss, 1967; Strauss & Corbin,
1998). This provides a clearer picture for a model of the phenomenon to be developed
and then tested using other methods.
However, the question still remains what is the correct sample size to anticipate? As
demonstrated in the preceding review it is a difficult question to answer. Some insight
might provided by a simple review of current studies that used grounded theory. One
hundred research articles from various disciplines that used grounded theory were
selected [1] from the seven year period, 2002 – 2008 (Appendix A). Grounded theory can
use a variety of data gathering techniques thus the articles chosen were restricted to
studies that used interviews. The average of all one hundred studies was 25; the range
was 5 to 114. Thirty-three of the studies used sample sizes between 20 and 30, thirty-two
used between 10 and 19, twenty-two used more than 31, twelve used under 10 and one
used more than one hundred (114). The review does not shed light on a specific number
but it does highlight several factors.
An analysis of the articles demonstrates the various factors that can be involved and how
they were applied. Expertise in the research area helped to facilitate a smaller sample size
as well achieving usable results (e.g. Jette, Grover & Keck, 2003). Using multiple
interviews with the same participant to gather more in-depth data led to a smaller sample
size (e.g. Troiano, 2003, or Lee, Woo & MacKenzie, 2002). The study that used more than
100 participants had a broad research question (e.g. Mason & Harris, 2006). The research
question dealt with the environmental factors that influenced market orientation using a
sample population across fifty organizations in various industries and of different sizes.
The authors state that since there was little known in the area it justified a large sample
population (Mason & Harris, 2006). However, for most research projects, conducting and
transcribing over one hundred interviews is very time consuming and increases the cost.
The authors may have been able to reduce the number of interviews by limiting some of
the controllable factors (e.g. industry).
The result of the analysis of the one hundred articles is that sample size for grounded
theory relies on the point of theoretical saturation (Glaser & Strauss, 1967; Locke, 2001,
Goulding, 2002, Strauss & Corbin, 1998). Researchers cannot make a judgment regarding
sample size until they are involved in data collection and analysis (Corbin & Strauss, 1998;
Glaser & Strauss, 1967). They must allow the data to dictate the sample size; therefore, it
is important to undertake data analysis during the data collection process. After each
interview the researcher or researchers should review the data and the emerging
themes. This review will help identify the point of theoretical saturation.

50
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
The literature review demonstrated that saturation normally occurs between 10 and 30
interviews. Although saturation might occur after the tenth interview, it is good practice
to test the level of saturation by conducting a few more interviews. Also these extra
interviews act as a form of validation of the patterns, concepts, categories, properties,
and dimensions that the researcher has developed from the previous interviews (Corbin
& Strauss, 1998). Thus, it would be wise to anticipate 30 interviews in order to facilitate
pattern, category, and dimension growth and saturation. It is only through the quality of
the data that meaningful and valid results are developed, so it is essential that the
researcher ensure that saturation has occurred (Corbin & Strauss, 1998; Glaser & Strauss,
1967).
Conclusion
The aim of this article was to provide an approximation of sample size requirements and
the reasoning for these requirements for a qualitative research method - grounded
theory. The answer to that question is that sample size will be dictated by theoretical
saturation and it can only be assessed during the data collection process. However, steps
can be taken to reduce the number of interviews required. Careful management of
research design factors such as focus of the research question, an awareness of the
nature or sensitivity of the phenomena, and a careful assessment of the ability,
experience or knowledge of the researcher can assist in reducing the number of the
interviews that are required. From the analysis of one hundred articles the recommended
is that researchers should plan on 30 interviews. This approximation will help guide
researchers in their initial project design stages to set out reasonable timeframes and
budget costs associated with data collection.
It was hoped that by utilizing a sample size of one hundred articles that it would capture
a representative sample of a commonly used qualitative method – grounded theory. A
limitation of this study is the fact that although the one hundred articles professed to use
grounded theory, it is uncertain how well the methodological procedures of grounded
theory were followed. However, the intention of the article is not to arrive at a definitive
number but at an approximation because, as stated, theoretical saturation will be the
final detriment of sample size when utilizing a grounded theory methodological
framework. Further it should be noted that although the arguments surrounding the
focus of the research question, sensitive nature of the phenomena and researcher’s
ability can be applied to other qualitative methods, the findings should not be
generalized to other qualitative research frameworks. Each method has its own nuances
that affect sample size and research should be undertaken to assess sample size
approximation. The aim of this project was to offer an approximation only to assist in the
design stage not to provide a set or given number.
Both qualitative and quantitative paradigms seek the truth. Therefore both traditions
strive to ensure their findings are generated from an appropriate sample size. Human
beings indeed present a complex system and when such a system is coupled with the
complexities of life, the understanding of how individuals interact is a daunting task
(Fornaciari & Lund Dean, 2001). By ensuring that researchers use an appropriate sample
size a step is taken towards greater validity and understanding of the complexities of life.

51
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
References
Auerbach, C. F. & Silverstein, L. B. 2003. Qualitative data: An introduction to coding and
analysis. New York: New York University Press.
Creswell, J. W. 1998. Qualitative inquiry and research design: choosing among five
traditions. Thousand Oaks CA.: Sage Publications.
Douglas, D. 2003. Grounded theories of management: A methodological review.
Management Research News, 26, pp. 44-60.
Fornaciari, C. J. & Lund Dean, K. 2001. Making the quantum leap: Lessons from physics on
studying spirituality and religion in the workplace. Journal of Organizational Change
Management, 14, pp. 335-51.
Glaser, B. G. & Strauss, A. 1967. The discovery of grounded theory: strategies for qualitative
research. Chicago, Ill: Aldine Pub. Co.
Goulding, C. 2002. Grounded theory: A practical guide for management, business and
market researchers. Thousand Oaks CA.: Sage Publications.
Huberman, A. M., & Miles M. B. 2002. Introduction. In A. M. Huberman & M. B. Miles
(Eds.) The Qualitative Researcher’s Companion, Thousand Oaks, CA: Sage Publications, pp.
37-64.
Jette, D. J., Grover, L. & Keck, C. P. 2003. A qualitative study of clinical decision making in
recommending discharge placement from the acute care setting. Physical Therapy, 83, pp.
224-36.
Jones, R. & Noble, G. 2007. Grounded theory and management research: A lack of integrity?.
Qualitative Research in Organizations and Management, 2, pp. 84-103.
Kwortnik, R. J. 2003. Clarifying “fuzzy” hospitality-management problems with depth
interviews and qualitative analysis. Cornell Hotel and Restaurant Administration Quarterly,
44, pp. 117-29.
LaRossa, R. 2005. Grounded theory methods and qualitative family research. Journal of
Marriage and Family, 67, pp. 837-857.
Lee, D. T. F., Woo, J. & Mackenzie, A. E. 2002. The cultural context of adjusting to nursing
home life: Chinese elders' perspectives. The Gerontologist, 42, pp. 667-75.
Locke, K. 2001. Grounded theory in management research. Thousand Oaks CA.: Sage
Publications.
Mason. K. J. & Harris, L. C. 2006. Market orientation emphases: an exploration of macro,
meso and micro drivers. Marketing Intelligence & Planning, pp. 24, 552-571.
Morse, J. 2000. Determining sample size. Qualitative Health Research, 10, pp. 3-5.

52
Thomson,
S. B. (201
1
).
Sample Size and Grounded
Theory.
JOAAG, Vol. 5. No. 1
Research Method/
Research Note
Sobal, J. 2001, Sample extensiveness in qualitative nutrition education research. Journal
of Nutrition Education, 33, pp. 184-192.
Strauss, A. & Corbin, J. 1998. Basics of Qualitative Research. Thousand Oaks, CA: Sage
Publications.
Troiano, P. F. 2003. College students and learning disability: Elements of self-style. Journal
of College Student Development, 44, pp. 404-419.