You are on page 1of 14

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/281981185

Does Sample Size Matter in Qualitative Research?: A Review of Qualitative


Interviews in is Research

Article  in  Journal of Computer Information Systems · September 2013


DOI: 10.1080/08874417.2013.11645667

CITATIONS READS

939 24,711

4 authors:

Bryan Marshall Peter Cardon


Georgia College University of Southern California
24 PUBLICATIONS   1,280 CITATIONS    73 PUBLICATIONS   2,010 CITATIONS   

SEE PROFILE SEE PROFILE

Amit Poddar Renee J. Fontenot


Salisbury University Georgia College
25 PUBLICATIONS   1,760 CITATIONS    20 PUBLICATIONS   1,519 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Does Sample Size Matter in Qualitative Research?: A Review of Qualitative Interviews in is Research View project

Best Practices in Diversity at Business Schools View project

All content following this page was uploaded by Peter Cardon on 04 November 2017.

The user has requested enhancement of the downloaded file.


Journal of Computer Information Systems

ISSN: 0887-4417 (Print) 2380-2057 (Online) Journal homepage: http://www.tandfonline.com/loi/ucis20

Does Sample Size Matter in Qualitative Research?:


A Review of Qualitative Interviews in is Research

Bryan Marshall, Peter Cardon, Amit Poddar & Renee Fontenot

To cite this article: Bryan Marshall, Peter Cardon, Amit Poddar & Renee Fontenot (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, DOI: 10.1080/08874417.2013.11645667

To link to this article: http://dx.doi.org/10.1080/08874417.2013.11645667

Published online: 10 Dec 2015.

Submit your article to this journal

Article views: 355

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at


http://www.tandfonline.com/action/journalInformation?journalCode=ucis20

Download by: [USC University of Southern California] Date: 07 December 2016, At: 16:50
DOES SAMPLE SIZE MATTER IN
QUALITATIVE RESEARCH?: A REVIEW OF
QUALITATIVE INTERVIEWS IN IS RESEARCH
BRYAN MARSHALL PETER CARDON AMIT PODDAR RENEE FONTENOT
Georgia College & University of Georgia College & Georgia College &
State University Southern California State University State University
United States United States United States United States

ABSTRACT to qualitative IS research (i.e., the Qualitative Research in IT


conference in New Zealand) and WebPages sponsored by IS
This study examines 83 IS qualitative studies in leading IS academic organizations that supply qualitative methodological
journals for the following purposes: (a) identifying the extent to resources on their websites (i.e., the Association of Information
which IS qualitative studies employ best practices of justifying System’s linked webpage http://www.qual.auckland.ac.nz/). Yet,
sample size; (b) identifying optimal ranges of interviews for like many disciplines, scant attention is paid to estimating sample
various types of qualitative research; and (c) identifying the extent size for qualitative interviews. In part, this may be due to the fact
to which cultural factors (such as journal of publication, number that qualitative research emerges from a paradigm of emergent
of authors, world region) impact sample size of interviews. Little design with a hesitation to estimate sample size at the often fluid
or no rigor for justifying sample size was shown for virtually and undefined initial stages of research. In this literature review,
all of the IS studies in this dataset. Furthermore, the number we focus first on theoretical arguments for estimating sample
of interviews conducted for qualitative studies is correlated size. Then, we discuss applied or practical reasons for estimating
with cultural factors, implying the subjective nature of sample sample size. Finally, we discuss best practices in justifying sample
size in qualitative IS studies. Recommendations are provided size and data saturation.
for minimally acceptable practices of justifying sample size of The concept data saturation (developed originally for
interviews in qualitative IS studies. grounded theory studies but applicable to all qualitative research
KEYWORDS: qualitative methodology, qualitative inter- that employs interviews as the primary data source) “entails
views, data saturation, sample size bringing new participants continually into the study until the data
set is complete, as indicated by data replication or redundancy.
INTRODUCTION In other words, saturation is reached when the researcher gathers
data to the point of diminishing returns, when nothing new is
Other than selecting a research topic and appropriate research being added” [2, 22]. Thus, estimating adequate sample size is
design, no other research task is more fundamental to creating directly related to the concept of saturation.
credible research than obtaining an adequate sample. Ensuring However, data saturation is an elusive concept and standard
that there is enough data is a precursor to credible analysis and in qualitative research since few concrete guidelines exist. As
reporting. Yet, rarely if ever do qualitative researchers justify Morse [23] stated, “Saturation is the key to excellent qualitative
the sample sizes of qualitative interviews. Furthermore, leading work . . . [but] there are no published guidelines or tests of ade-
qualitative research methodologists provide few concrete quacy for estimating the sample size required to reach satur-
guidelines for estimating sample size. ation.” Furthermore, many qualitative methodologists recognize
In this study, we explore how well researchers have justified the sloppy fashion by which sample size is described in qualita-
their sample sizes for qualitative interviews in leading IS journals. tive studies. Onwuegbuzie and Leech [29] stated, “Many
We began the study with the hope of establishing evidence-based qualitative researchers seemingly select the size of their samples
guidelines for researchers who intend to use qualitative interviews in an arbitrary fashion” (pp. 115-116). Charmaz [4] further
in their research. Furthermore, our hope was that these guidelines stated:
would lead to publishable and impactful IS research. In the
Findings and Discussion sections of the paper, we demonstrate Often, researchers invoke the criterion of saturation to
the poor compliance with basic procedures for rigor in sample justify small samples — very small samples with thin data.
size. We conclude with guidelines for standard practices related to Such justifications diminish the credibility of grounded
justifying sample size that should be employed in all qualitative- theory. . . . Claims of saturation often reflect rationalization
interview-based studies. We also discuss what we consider a crisis more than reason, and these claims raise questions.
of rigor related to sample sizes in qualitative IS research. What stands as a category? Is it conceptual? Is it useful?
Developed by whose criteria? All these questions add up
LITERATURE REVIEW to the big question: What stands as adequate research?
(p. 528)
Abundant qualitative research exists in the IS field, including
a variety of reviews and recommendations tailored specifically Most qualitative methodologists openly recognize the lack
to the use of qualitative methodology in the IS field [1, 9, 16, of standards for sample size. At the same time, some qualitative
17, 18, 19, 26, 40]. There have even been conferences devoted methodologists are not troubled by the lack of guidelines, even

Fall 2013 Journal of Computer Information Systems 11


considering the vague nature of sample size guidelines as a coverage of the phenomenon given the purpose of the study and
reflection of the qualitative orientation to research. This orientation stakeholder interests.” (p. 246) Patton [31] further stated:
is not only theoretical but also psychologically fitting for people
with more tolerance for ambiguity. For example, in the Sample At the beginning, for planning and budgetary purposes,
Size section of his work on qualitative research methodology, one specifies a minimum expected sample size and builds
Patton [31] explains, rationale for that minimum, as well as criteria that would
alert the research to inadequacies in the original sampling
Qualitative inquiry is rife with ambiguities. There are approach and/or size. In the end, sample size adequacy, like
purposeful strategies instead of methodological rules. all aspects of research, is subject to peer review, consensual
There are inquiry approaches instead of statistical formulas. validation, and judgment. (p. 246)
Qualitative inquiry seems to work best for people with a
high tolerance for ambiguity. . . . Nowhere is this ambiguity Many qualitative methodologists claim the time and mentally
clearer than in the matter of sample size. . . . There are no exhaustive nature of qualitative work compared to other forms of
rules for sample size in qualitative inquiry. Sample size research. Yin [45] stated regarding case study research:
depends on what you want to know, the purpose of the
inquiry, what’s at stake, what will be useful, what will have In actuality, the demands of a case study on your intellect,
credibility, and what can be done with available time and ego, and emotions are far greater than those of any other
resources. (pp. 242-243) research method. This is because the data collection
procedures are not routinized. . . . During data collection,
While qualitative methodologists are unlikely to agree on only a more experienced investigator will be able to take
exact sample sizes needed for qualitative studies, they generally advantage of unexpected opportunities rather than being
agree that a number of factors can affect the number of interviews trapped by them — and also will exercise sufficient care
needed to achieve saturation. In addition to the nature and scope against potentially biased procedures. (p. 68)
of the researcher, some other factors that can influence sample
size needed to reach saturation include quality of interviews, Guest, Bunce, and Johnson [15] further emphasized the need for
number of interviews per participant, sampling procedures, and numerical targets for sample sizes of interviews:
researcher experience [21, 24, 31, 32, 34, 35].
Estimating and justifying sample size of interviews has Our experience, however, tells us that it is precisely a general,
more than theoretical significance. Fundamentally, qualitative numerical guideline that is most needed, particularly in the
researchers face a number of unique challenges when proposing, applied research sector. Individuals designing research —
conducting, and getting their research published. Fundamentally, lay and experts alike — need to know how many interviews
there is still strong resistance among many scholars to qualitative they should budget for and write into their protocol, before
research. Although the standing of qualitative work has improved they enter the field. This article is in response to this need,
over the past few decades, qualitative work is still viewed as and we hope it provides an evidence-based foundation on
less rigorous research by many non-qualitative researchers which subsequent researchers can expand. (p. 60)
[8]. Furthermore, many qualitative research techniques are
misunderstood by non-qualitative researchers. This leads to We have cited a number of scholars who describe the
a variety of political problems, including difficulty in getting importance of estimating and justifying sample size based on
qualitative proposals approved and getting qualitative work practical reasons. We would add that we consider the lack of more
published. Many scholars also believe that qualitative research is concrete guidelines acts as a barrier to many scholars in considering
more time consuming and mentally challenging. Therefore, it can qualitative research. When Yin [45] states that “procedures are
be a high-resource, high-risk, high-time commitment research not routinized . . . [and that] only a more experienced investigator
activity in an academic environment that is often driven by short- will be able to take advantage of unexpected opportunities rather
term productivity demands. For these many reasons, it is essential than being trapped by them” (p. 68), this deters scholars who,
that qualitative researchers identify ways of most efficiently despite interest in qualitative techniques, prefer some routine
accomplishing their research objectives. Getting the right amount and less ambiguity. Furthermore, when Patton [31] states that
of data through interviews can accomplish this objective. “qualitative inquiry seems to work best for people with a high
Politically, qualitative researchers are generally in a minority tolerance for ambiguity . . . [and that] nowhere is this ambiguity
position. Doctoral committees and editorial review boards clearer than in the matter of sample size” (pp. 242-243), he may
typically contain a majority of quantitative-oriented researchers. be unintentionally excluding a large set of scholars who are
Therefore, qualitative researchers need to mold research de- sympathetic to qualitative research but who need more structured
sign, analysis, and reporting to get political support. Many guidelines for rigor, including for estimating sample size. These
reviews of qualitative research proposals are conducted with barriers to entry for researchers with tighter expectations of rigor
scholars with little or no knowledge of proper qualitative are a practical problem for all qualitative researchers because
procedures. In some cases, antagonism towards qualitative work by inadvertently discouraging other scholars from participating
exists [5, 25, 30]. in qualitative research, the minority position of qualitative
Regarding the many political and resource constraints facing researchers is perpetuated, thus perpetuating the political and
qualitative researchers, Patton [31] stated: “Sampling to the point consensus-gaining activities that qualitative researchers must
of redundancy is an ideal, one that works best for basic research, engage in more so than quantitative researchers.
unlimited timelines, and unconstrained resources. The solution is In this section, we describe three methods that can be used
judgment and negotiation. I recommend that qualitative sampling to justify sample size of interviews in qualitative research.
designs specify minimum samples based on expected reasonable The first and second are external justifications — they depend

12 Journal of Computer Information Systems Fall 2013


on other scholars. The first method is to cite recommendations reached” (p. 118). In this situation, researchers estimate and/or
by qualitative methodologists. The second method is to act on justify sample size by citing similar studies that claimed data
precedent by citing sample sizes used in studies with similar saturation at certain points.
research problems and designs [38]. The final and third method The third method of justifying sample size is through
is internal justification. It involves statistical demonstration of statistical demonstration of saturation within a dataset. We’re
saturation within a dataset. aware of few research efforts to identify appropriate sample size
Some qualitative research methodologists present general ranges. One exception is the work of Guest, Bunce, and Johnson
guidelines for sample size of interviews [36]. These guidelines [15] regarding the issue of theoretical saturation. Based on their
vary from methodologist to methodologist and sometimes the review of recommendations of qualitative methodologists, they
same methodologist has provided different ranges at different found that nearly all recommend achieving theoretical saturation,
points in time; however, there is substantial overlap in the various yet stated the following:
recommended ranges. For example, in grounded theory studies,
Creswell [6] recommends at least 20 to 30 interviewees. Denzin They [qualitative methodologists] did a poor job of
and Lincoln [8] recommend 30 to 50 interviews. Morse [24] operationalizing the concept of saturation, providing no
recommends 20 to 30 interviewees with 2 to 3 interviews per description of how saturation might be determined and
person. In 1994, she had recommended 30 to 50 interviews and/ no practical guidelines for estimating sample sizes for
or observations. For phenomenological studies, ranges include purposively sampled interviews. This dearth led us to carry
approximately 6 [7], 6-8 [20], and 6-10 [24]. Case studies are out another search through the social science and behavioral
among the most difficult types of qualitative research to classify. science literature to see if, in fact, any generalizable
Yin [44] recommends at least six sources of evidence. Creswell [6] recommendations exist regarding nonprobabilistic sample
recommends no more than 4 or 5 cases. He further recommends sizes. After reviewing twenty-four research methods books
3 to 5 interviewees per case study. One problem is that when and seven databases, our suspicions were confirmed; very
qualitative methodologists do in fact present number ranges for little headway has been made in this regard. (p. 60)
appropriate sample size, they fail to explain any rationale [29].
The second method of defending sample size is by precedent. Thus, they set out to develop evidence-based guidelines from
Onwuegbuzie & Leech [29] stated, “We recommend that before their dataset of interviews of sixty women in Nigeria and Ghana
deciding on an appropriate sample size, qualitative researchers in a phenomenological study of social desirability bias and self-
should consider identifying a corpus of interpretive studies that reported sexual behavior. They followed coding procedures of
used the same design as in the proposed study (e.g., grounded Glaser and Strauss [13]. Based on themes developed in their
theory, ethnography) and wherein the data saturation was codebooks, they concluded that of 109 content-driven codes, 80

FIGURE 1 — Data Saturation in a Phenomenological Study Note. Based on Guest, Bunce, and Johnson

Fall 2013 Journal of Computer Information Systems 13


(73%) were identified within the first six interviews transcripts; qualitative research (i.e., grounded theory, phenomenology, case
100 (92%) were identified within the next six interview transcripts; study) would tend to cluster around an optimal range of interviews
the final 8 codes (to reach 100% of all codes) were identified by for several reasons. First, for theoretical reasons and assuming
completion of the thirtieth interview transcript (see Figure 1). In that researchers verify data saturation during the coding process
terms of thematic code prevalence, they calculated a Cronbach’s with statistical techniques, we would expect that most studies
alpha to measure the reliability of code frequency distribution would reach theoretical saturation most frequently in roughly
as the analysis progressed. Generally, .70 is considered an the same range. Generally, to justify data saturation would
acceptable measure. They showed that within 12 interviews, require a researcher to conduct several interviews past that point
the Cronbach’s alpha was .70; after 18 interviews, .79; after 24 (to indicate that the dataset was indeed becoming redundant).
interviews, .85; after 30 interviews, .88; after 36 interviews, Furthermore, we would expect that this range would be influenced
.88; after 42 interviews, .89; after 48 interviews, .90; after 54 by the recommendations of qualitative methodologists. We would
interviews, .91; after 60 interviews, .93. They concluded that this also expect that study quality to be highest at about the point of
measure was reliable early in the process and improved at ever data saturation. In other words, the quality of the study would
decreasing rates. They concluded that most of the data saturation increase until data saturation is reached but diminish afterwards.
had occurred by 12 interviews. They state the following about As illustrated in Figure 1, there are rapidly diminishing returns
the questions researchers should ask when planning sample size: once the data is saturated. However, exceeding the point of data
“The question we pose, however, frames the discussion differently saturation is not necessarily a harmless activity. In fact, gathering
and asks, ‘Given x analyst(s) qualities, y analytic strategy, and z too much data can impair researchers from the deep, rich
objectives(s), what is the fewest number of interviews needed to analysis of the data that is a hallmark and central to the purpose
have a solid understanding of a given phenomenon?’” (p. 77). of qualitative research [31]. Furthermore, for most researchers,
Based on our review of literature about justifying sample moving far past the point of saturation would devour limited time
size of qualitative interviews, we would hypothesize that several and resources. Therefore, we would expect the quality imperative
relationships would emerge when qualitative researchers follow to reign in efforts to gather too much data.
best practices for justifying sample size, as depicted in Figure 2.
In the figure we have assumed that there is a point where data is METHODS
most often saturated — in this case about 30 interviews (a rough
midpoint of suggested ranges for grounded theory methodologists). The primary purposes of our research were the following: (a)
This assumption is obviously variable and dependent on the type identify the extent to which IS qualitative studies employ best
and scope of study as well as the skill set of the researcher/s. practices of justifying sample size; (b) identify optimal ranges of
The larger point is that we would assume that certain types of interviews for various types of qualitative research; (c) identify

FIGURE 2 — Hypothesized Relationships with Sample Size of Qualitative Interviews

14 Journal of Computer Information Systems Fall 2013


the extent to which cultural variables (i.e., journal, author, region) are the primary data source, since it is through this method that
impact sample size of interviews; and (d) identify the relationship the researcher can best access the interpretations that participants
between sample size of interviews and impact of the research. have regarding the actions and events which have or are taking
Thus, we had the following broad research questions: place” (p. 78).
We identified articles that employed interviews as the primary
s$O )3 RESEARCH STUDIES CONFORM TO BEST PRACTICES FOR data source in the following journals: MIS Quarterly, Informa-
justification of sample size? tion Systems Journal, Information Systems Research, Communi-
s(OWMANYQUALITATIVEINTERVIEWSAREENOUGHFORVARIOUS cations of the ACM, and Journal of Management Information
types of qualitative research? Systems. We selected these five journals due to the high quality
s$O CULTURAL VARIABLES IE JOURNAL AUTHOR REGION of the journals and accessibility through our library systems.
influence the sample size of interviews? Four of the five selected journals are top five IS journals accord-
ing to average rankings on the Association for Information
We are aware of few efforts to examine qualitative studies as Systems website, which averages rankings of nine studies of IS
an evidence-based guide to appropriate sample size in qualita- journal quality.
tive studies. One example of such an effort is Thomson’s [39] We gathered several pieces of information about the
review of fifty ground theory articles during the 2002-2004 interviews. After all, “Sampling involves more than just the
time period. Sample size ranged from 5 to 350 with an aver- number of participants included in the study; sampling is a
age of 31. By taking the 350-interview study out of the sample, process that incorporates the number of participants, the number
the average number of interviews was 24. Based on his study, of contacts with each participant, and the length of each contact.
he made several conclusions: (a) small sample size studies It is important for researchers to consider how much contact
generally involved more contact time with each interviewee will be required in order to reach saturation” [29]. Therefore,
(longer interviews and/or repeated interviews); (b) theoretical we gathered the number of interviewees, interviews, and length
saturation generally occurs between 10 and 30 interviews; and of interviews. By multiplying the number of interviews with
(c) once a researcher believes saturation has occurred, he/she the length of interviews, we established total contact time. In
should conduct several additional interviews to test whether cases where authors simply stated a range of interview lengths,
existing themes and categories are sufficient. Based on the we average the range (e.g., 60 to 90 minutes was average to 75
dilemmas of estimating needed interviews beforehand, he stated: minutes).
“Thus, it would be wise to anticipate 30 interviews in order to We wanted to examine the impact of rigor of sample size
facilitate pattern, category, and dimension growth and satura- justification with study quality. We decided to use impact factor
tion. It is only through the quality of the data that meaningful as a rough estimate of study quality. This is by no means a perfect
and valid results are developed, so it is essential that the re- measure of study quality, but it is a decent measure of how
searcher ensure that saturation has occurred.” well accepted a study is. To determine impact we used several
A study that focused specifically on rigor in the IS field was methods. First, we used citations and impact factors developed
Dubé and Paré’s [9] study of 183 IS case studies in seven IS by Web of Science®. For each article in the database (available
journals from 1990 to 1999. They examined the degree to which for all articles in our sample with the exception of Journal of
these studies met standards of rigor as posited by case study Management Information Systems), we were able to obtain the
methodologists. They found that “a large portion of them have number of articles in journals in the ISI Web of Knowledge index
actually ignored the state of the art of case research methods that cited the article. We calculated five-year impact factors by
that have been readily available to them” (p. 599). They found calculating the average number of citations during the five years
that semi-structured interviews were the primary data collection immediately following publication. Second, we used Google
method in 88% of the articles. Only 13% of articles described Scholar to identify the number of citations for each article. For
the sampling strategy. Fewer than 38% included the number of Google impacts, we calculated average annual citations for all
interviewees and only 24% described the number of interviews articles that had been published for at least five years.
conducted. “In short, the apparent lack of information about
the sampling strategy in positivist case studies might prohibit FINDINGS
the reader from understanding the limits of the conclusions
that are drawn from such research” (p. 614). They found that As far as justifying sample size for interviews, there was
in addition to qualitative interviews, other major data sources little if any defense based on the three best practices for doing
used the case studies were existing documentation (64% of so. First, no studies cited qualitative methodologists for an
cases), direct observation (32% of cases), and quantitative surveys appropriate sample size. Several single case studies cited Yin [44]
(27% of cases). in defense of using a single case but went no further in describing
We patterned our study primarily on Thomson’s [39] study of why they employed the number of interviews they conducted.
grounded theory articles, yet we wanted to expand our sample to One multiple-case study cited Eisenhardt [10] in the selection
studies that employed qualitative interviews (not just grounded of ten cases; yet, provided no additional detail about reasons for
theory studies) but restrict the sample to IS studies. While it is selecting the number of interviews for each case [3]. Many of the
true that many archival documents, observation, and other sources studies included references to theoretical saturation as described
of data are included in qualitative studies, particularly cases, by Glaser and Strauss [13]. Surprisingly, even when mentioning
qualitative interviews are the primary data source in the vast the importance of saturation, none of these studies provided a
majority of IS positivist cases studies, as confirmed by Dubé and defense of the size of their sample. As examples, Nissen [28]
Paré [9]. This is not different for the interpretive approach to case stated: “[We conducted] more than 200 [interviews] conducted
studies. As Walsh [40] stated, “With respect to interpretive case over a three-month period. The number, time, and scope of
studies as an outside observer, it can be argued that interviews interviews continue until theoretical saturation is reached” (p.

Fall 2013 Journal of Computer Information Systems 15


235); yet, they never explained how or when saturation was pants who have been involved in particular issues and policies”
achieved. Garud and Kumaraswamy [12] cited Glaser and Strauss (p. 54). Goulielmos [14] stated, “No new concepts were devel-
[13] and explained the importance of “iterating between theory oped with the completion of data completion” (p. 366). These
until a stage of theoretical saturation is reached” (p. 15); yet, types of statements were rare, and in each instance, presented
similarly never mentioned at what point this was reached in their in one or two sentences with no statistical demonstration of
study. Watson-Manheim and Bélanger [41] advocated “reaching sample size.
theoretical saturation when possible” (p. 271); yet, never discussed We provide various descriptive statistics that show ranges of
how their sample reached saturation. In short, none of the studies number of interviews for various types of research methodolo-
cited qualitative methodologists in defense of their sample size. gies. In Table 1, we group these statistics by the following types
Many of the studies invoked the concept of saturation but never of research studies: grounded theory, single case, multiple case
described a point at which that was expected to or did happen. study, other types of qualitative study, and mixed design (quali-
For the second method of defending sample size, only one tative and quantitative) studies. The final two columns show
study out of the 83 studies cited a prior qualitative work as the statistics for the top 25 percentile in impact for articles that
basis for sample size when addressing a similar problem. Ryan were published in 2005 or before, thus focusing on impacts that
and Harrison [33] stated, “Other IT researchers have used a have extended over a minimum of five years. Google top per-
similar number of interviews to explore complex issues and build formers refer to average annual citation counts for articles older
theory” (p. 17) and then cited the work of Niederman, Beise, and than five years. Top performers refer to five-year impact factors
Beranek [27] as a basis for their sample size. of articles older than five years according to the Social Sciences
For the third method of defending sample size, none of the Index. Institution region refers to the location of the first author’s
studies cited statistical evidence of saturation. A few studies institution and the regions are North America (United States
made vague statements about reaching saturation near the end of and Canada), Europe (including Australia and New Zealand), and
the interviews. Tallon and Kraemer [37] stated, “We conducted Asia. Impact factors are displayed in quartiles since with relatively
interviews until we reached theoretical saturation — that is, the small sample sizes, averages are easily skewed by extreme points
last few interviews did not provide new insights — making the in the dataset. Since grounded theory, single case, and multiple
sample size appropriate for this study” (p. 145). Fort, Larco, and case research designs are the dominant forms of qualitative
Bruckman [11] stated, “When themes began arising over and research, we report only on those types of research throughout
over in these interviews, we followed up by recruiting partici- the remainder of the paper.

TABLE 1. Descriptive Statistics of Included Studies


Grounded Multiple Google Top Top
Theory Single Case Case Other Mixed All Performers Performers
Time Period n % n % n % n % n % n % n % n %
1980s 0 0% 0 0% 0 0% 3 18% 1 6% 3 4% 0 0% 0 0%
1990-1994 0 0% 2 6% 0 0% 1 6% 0 0% 3 4% 1 8% 0 0%
1995-1999 3 15% 4 12% 3 13% 1 6% 2 11% 9 11% 4 33% 1 13%
2000-2004 8 40% 12 35% 9 39% 6 35% 7 39% 31 37% 7 58% 7 88%
2005-2009 9 45% 16 47% 11 48% 6 35% 8 44% 37 45% 0 — 0 —
Journal
CACM 4 20% 2 6% 1 4% 7 41% 3 17% 13 16% 0 0% 1 13%
ISJ 2 10% 11 32% 7 30% 2 12% 6 33% 22 27% 0 0% 0 0%
ISR 3 15% 1 3% 4 17% 0 0% 1 6% 6 7% 2 17% 3 38%
JMIS 8 40% 16 47% 5 22% 3 18% 5 28% 26 31% 5 42% 0 0%
MISQ 3 15% 4 12% 6 26% 5 29% 3 17% 16 19% 5 42% 4 50%
Institution Region
USA/CA 14 70% 17 50% 14 61% 13 76% 8 44% 49 59% 9 75% 6 75%
Europe 5 25% 13 38% 6 26% 4 24% 7 39% 26 31% 2 17% 1 13%
Asia 1 5% 4 12% 3 13% 0 0% 3 17% 8 10% 1 8% 1 13%
Number of Authors
1 5 25% 6 18% 4 17% 6 35% 6 33% 17 20% 2 17% 1 13%
2 11 55% 16 47% 14 61% 4 24% 6 33% 39 47% 8 67% 6 75%
3 or more 4 20% 12 35% 5 22% 7 41% 6 33% 27 33% 2 17% 1 13%
Impact Factor
Minimum 0.0 2.0 0.0 0.0 0.0 0.0 0.0 20.0
Quartile 1 6.8 6.5 5.0 3.0 3.5 4.5 10.5 26.8
Quartile 2 17.5 9.0 16.0 5.0 5.0 9.0 27.0 28.5
Quartile 3 27.5 24.5 20.0 11.0 8.5 18.5 29.5 30.3
Maximum 48.0 48.0 30.0 28.0 27.0 48.0 48.0 48.0
Total 20 34 23 17 18 83 12 8

16 Journal of Computer Information Systems Fall 2013


In Table 2, we present the number of interviews per study Journal, 69% of the articles contained 30 or fewer interviews,
broken down by journal, institution region, and number of compared to 66% in Information Systems Research, 56% in
authors. Generally speaking, some journals contained articles Journal of Management Information Systems, and 38% in MIS
with far more interviews per study. For example, 50% of articles Quarterly. European and Asian authors were far more likely to
in MIS Quarterly contained more than 40 interviews, compared include fewer interviews in their studies with 77% of European
to 34% in Journal of Management Information Systems, 23% authors including fewer than 30 interviews, compared to 71%
in Information Systems Journal, 17% in Information Systems for Asian authors, and 50% for American authors. Studies that
Research, and 15% in Communications of the ACM. On the other included more authors similarly contained fewer interviews. In
hand, some journals contained articles with far fewer interviews. approximately 76% of studies authored by three or more authors
In Communications of the ACM and Information Systems there were fewer than 30 interviews compared to 52% and 47%

TABLE 2. Number of Interviews per Study by Journal, Region, and Authors


Under 10 11 to 20 21 to 30 31 to 40 41 to 50 51 to 100 over 100
Journal n % n % n % n % n % n % n %
CACM 2 15% 0 0% 7 54% 2 15% 0 0% 2 15% 0 0%
ISJ 5 23% 7 32% 3 14% 2 9% 4 18% 0 0% 1 5%
ISR 0 0% 2 33% 2 33% 1 17% 0 0% 1 17% 0 0%
JMIS 3 13% 7 30% 3 13% 2 9% 3 13% 4 17% 1 4%
MISQ 1 6% 3 19% 2 13% 2 13% 2 13% 5 31% 1 6%
Institution Region
USA/CA 4 8% 10 20% 11 22% 8 16% 3 6% 11 22% 2 4%
Europe 6 27% 8 36% 3 14% 0 0% 3 14% 1 5% 1 5%
Asia 1 14% 1 14% 3 43% 0 0% 2 29% 0 0% 0 0%
Number of Authors
1 1 6% 6 35% 1 6% 3 18% 3 18% 2 12% 1 6%
2 5 13% 5 13% 10 26% 3 8% 5 13% 8 21% 2 5%
3+ 5 20% 8 32% 6 24% 3 12% 1 4% 2 8% 0 0%
Note. Percentages are in rows.

FIGURE 3. Number of Interviews Based on Number of Authors

Fall 2013 Journal of Computer Information Systems 17


in double-authored and single-authored studies, respectively. We containing fewer than 20 interviews) or large samples sizes (30%
depict this author relationship in Figure 3 to show that whereas containing more than 40 interviews). The majority (57%) of
there is little pattern to the number of interviews in single-authored multiple case studies contains 20 to 50 interviews; however, there
and double-authored studies, those studies with three or more is major variability.
authors tend to crest around twenty interviews and gradually fall In Table 4, we present the number of cases in multiple case
to around fifty interviews with quite uncommon cases of studies studies. Generally, most of these studies contain just 2 or 3 cases
with more than fifty interviews. (43%) or 8 or more cases (43%). Few studies contain between 4
In Table 3, we present the number of interviewees, interviews, and 7 cases (13%). This pattern is similar to grounded theory and
and contact time (in total hours per study) for grounded theory, single case designs in that sample sizes tend to cluster around
single case, multiple case, Google top performers, and top small or large sample sizes. And surprisingly, this is counter to
performers. The differential between minimums and maximums the recommendations of some qualitative methodologists, such
is extremely large. For example, in terms of total
contact time the differential ranges on order of 17
TABLE 3. Interviewees, Interviews, and Contact Time by Research Design
to 40 times. Even when limiting analysis of the
middle fifty percentile, the range for grounded Grounded Single Multiple Google Top Top
theory, single case, and multiple case studies is on Theory Case Case Performers Performers
an order of 3 to 4.5 times: grounded theory ranges
Interviewees
from 16 to 70 contact hours for interviews; single
Minimum 6 4 10 7 12
case ranges from 10 to 46 hours; and multiple case
Quartile 1 15 13 22 19 20
ranges from 21 to 69 contact hours. Ranges tighten
Quartile 2 27 24 39 29 24
for those studies with highest impacts: Google top
Quartile 3 59 43 45 68 40
performers range from 31 to 71 contact hours and
Maximum 200 200 74 105 105
Social Science Index top performers range from 23
to 39 contact hours. Interviews
In Figure 4, we depict trends for sample size Minimum 6 4 10 7 12
of interviews for grounded theory, single case, and Quartile 1 17 13 22 19 20
multiple case studies. The trend line for grounded Quartile 2 29 23 40 29 24
theory studies is particularly surprising since there Quartile 3 59 45 49 68 40
are more concrete suggestions for grounded theory Maximum 200 200 74 127 127
studies in terms of sample size (generally ranging Contact Time (hrs)
in the 20 to 40 range). The IS grounded theory Minimum 5.5 4.0 6.0 5.5 12.0
studies we examined tended to be on either side of Quartile 1 15.8 10.0 21.2 31.2 23.0
this range with 35% of these studies containing less Quartile 2 37.3 28.0 38.8 39.0 37.3
than 20 interviews and 40% containing more than Quartile 3 70.0 46.0 68.8 71.0 38.8
40 interviews. Similarly, single case studies exhibit Maximum 222.3 222.3 100.5 222.3 222.3
a similar trend with either small sample sizes (45%

FIGURE 4. Interviews per Study by Research Design

18 Journal of Computer Information Systems Fall 2013


TABLE 4. as Creswell, who recommend not exceeding five or six cases in
Number of Cases in Multiple Case Research Designs multiple case designs.
Finally, in Figures 5 and 6 we show scatter plots with trend
n %
lines with the relationships between number of interviews and
Min # Cases 2 average annual Google Scholar citations for articles that were
Average Number of Cases 7 published five or more years ago. We present these figures
Median Number of Cases 5 cautiously due to the small sample sizes. Nonetheless, we find
Max # Cases 34 it interesting that there are trends showing maximum impact
# Cases per Study around 25-30 interviews for grounded theory studies and 15-25
2 7 30.4% interviews for single case studies.
3 3 13.0%
4 1 4.3% DISCUSSION AND RECOMMENDATIONS
5 2 8.7%
6 0 0.0% Without justification of sample size, the implication is that
7 0 0.0% sample size is arbitrary and thus inconsequential. In the studies
8 2 8.7% we examined, there is no apparent effort to justify sample size
9 2 8.7% by citing the recommendations of qualitative methodologists,
10 or more 6 26.1% acting on the precedent of other studies with similar designs
and research problems, or by demonstrating statistically that the

FIGURE 5.
Number of
Interviews
and Google
Scholar Annual
Citations for
Grounded
Theory Studies

FIGURE 6.
Number of
Interviews
and Google
Scholar Annual
Citations for
Single Case
Studies

Fall 2013 Journal of Computer Information Systems 19


dataset collected had become saturated. Rather, IS researchers oriented justification for sample size. After all, the best and most
made superficial references to data saturation or altogether rigorous justification for sample size of interviews does not
omitted justification for sample size. Since justifying sample emerge from the steps a researcher takes in collecting the data
size is evidence that the dataset is sufficient to address research (process-driven), it emerges with statistical demonstration of
problems, this indicates an indifference to rigor that would be redundancy in codes (results-driven).
unacceptable in quantitative research and should likewise be Although there are major limitations in each study in terms
unacceptable in qualitative research. of justifying sample size, we do think that we can infer some
We anticipated that we would be able to provide evidence- collective wisdom from the studies that provide guidelines for
based guidelines for the number of interviews needed for qualitative IS studies. We recommend the following:
various types of research designs. Ultimately, we showed the Grounded theory qualitative studies should generally include
extreme variation in sample size in all research designs. Rather between 20 and 30 interviews. We based this recommendation
than clustering in a mid-range of ideal number of interviews, on several pieces of information from the study. Assuming that
grounded theory and single case studies tended to have many top performers are articles that are most highly respected, we
studies with small sample sizes and many studies with large think that the minimum number of articles, based on precedent,
sample sizes. Furthermore, there were few studies in suggested should be the minimum standard in the majority of top per-
ranges (mid-ranges) by qualitative methodologists. For studies formers (in this case, we view this majority as the Quartile 1
with small sample sizes (for example, grounded theory studies figure as displayed in Table 3 for top performers and Google
of 20 or fewer interviews) and without justification of sample top performers). At the same time, qualitative researchers
size, it would be easy to question whether theoretical saturation should assume that too many interviews can be counterproduc-
had been reached. For studies with large sample sizes (for tive. As a result, we think the maximum should be where addi-
example, grounded theory studies of 40 or more interviews), tional interviews fail to produce substantial new insight. We see
it would be easy to question whether the researchers had been no evidence that studies over 30 interviews yielded significantly
able to devote sufficient attention to analyzing and reporting more impact. For example, Figure 5 shows that the average
in-depth, rich content with such a voluminous dataset. number of annual Google citations essentially reached a plateau
As far as the cultural influence (values and norms of groups around 30 interviews. We also think exceeding 30 interviews
and institutions), it is evident that culture does play a role. Some generally defies the wisdom when multiple authors work toge-
journals tend to publish articles with much larger sample sizes ther. When three or more authors work together, they view
(such as MISQ), some regions (based on institution of first author) saturation as occurring with 30 or fewer interviews in the vast
likewise include more interviews (such as North America), and majority (76%) of studies (see Figure 4). We particularly cau-
larger groups of researchers (three or more) tend to conduct tion against grounded theory studies of over 40 interviews.
fewer interviews. The fact that there is such variation highlights The collective judgment in our study seems to indicate this as
the subjective nature of determining sample size. Given this well since 60 percent of all grounded theory studies contained
variation, the need for evidence-based guidelines for sample size fewer than 40 interviews (see Figure 4). We expect further
of interviews is further needed. refinement to our recommended range in years to come as IS
We consider the lack of rigor in justifying sample size a researchers report the techniques they use to reach data satur-
major crisis to qualitative IS researchers. At a minimum level, ation (as outlined in the prior recommendation).
we recommend that IS researchers use each of the three best Single case studies should generally contain 15 to 30
practices for justifying sample size. The most critical best prac- interviews. The extreme variation in practices for single case
tice is statistical demonstration of data saturation since this studies makes a recommendation challenging. What is clear is
provides internal support for the value of the dataset and the that 69 percent of all qualitative IS studies sampled for this study
analysis and reporting built on the dataset. We consider the employed fewer than 30 interviews. We think it would be rare that
second most important best practice that of citing other similar additional interviews would be a wise time investment. Further,
studies that have adopted similar designs with similar research of top Google performers, the highest average impacts fell in the
problems. 15 to 30 interview range (see Figure 5).
There are many benefits to adopting these best practices. Qualitative researchers should examine the expectations
First, they add to the credibility of the research. Second, they of their intended journal outlets based on history and culture.
have the potential of saving significant amounts of time. Since It appears that some journals may have expectations for higher
each interview creates enormous time commitments (such as numbers of interviews. For example, qualitative studies in MIS
establishing relationships of trust, transcribing, coding), focusing Quarterly have far higher numbers of qualitative interviews per
on getting the right amount of data can save hundreds of hours. study. Furthermore, it appears that American researchers tend to
Third, we believe that adopting more rigorous standards of conduct more interviews. We do not know a reason for this. Aside
qualitative research would enhance the reputation of qualitative from the theoretical and statistical elements of justifying sample
research generally and even make this type of research more size, qualitative researchers should recognize these apparent
appealing to quantitative researchers. cultural judgments of data saturation and adjust their sample sizes
None of the three best practices described in this article are accordingly.
new nor are they controversial among most qualitative researchers Replication studies should further examine the impacts of
— yet, few if any IS qualitative researchers are employing them. culture and study design. We think this study shows clearly that
As we reviewed the many qualitative articles for this study, we culture and study design affect judgments about data saturation.
concluded that there is a heavy emphasis on process in methods Future studies that specifically addressed to what extent, how, and
sections. Qualitative researchers carefully detailed all of the many why cultural factors affect judgments of data saturation would
steps they took in the data collection and analysis process. This be particularly helpful. We also think future studies are needed
process-heavy orientation in terms of methods omits the results- to examine ideal ranges of qualitative interviews, particularly in

20 Journal of Computer Information Systems Fall 2013


multiple case studies. We recommend expanding the inclusion positivist case research: Current practices, trends, and
of other top IS journals for such studies, including Association recommendations,” MIS Quarterly (27:4), 2003, 597-635.
of Information Systems, Journal of Information Technology, [10] Eisenhardt, K. M. “Building theories from case study
Journal of Strategic Information Systems, European Journal of research,” Academy of Management Review, (14:4), 1989,
Information Systems, and Journal of AIS. 532-550.
[11] Forte, A., Larco V., & Bruckman, A. “Decentralization in
SUMMARY Wikipedia governance,” Journal of Management Information
Systems (26:1), 2009, 49-72.
In this paper, we have addressed the problem of estimating [12] Garud, R., & Kumaraswamy, A. “Vicious and virtuous
and justifying sample size of qualitative interviews. By examining circles in the management of knowledge: The case of Infosys
83 IS qualitative studies in leading IS journals, we have shown Technologies,” MIS Quarterly (29:1), 2005, 9-33.
that there is little rigor in justifying sample size. Furthermore, [13] Glaser, B. G., & Strauss A. L. The Discovery of Grounded
the number of interviews conducted for qualitative studies is Theory. Aldine, New York, NY, 1967.
correlated with cultural factors (such as journal of publication, [14] Goulielmos, M. “Systems development approach: tran-
number of authors, world region), implying the subjective nature scending methodology,” Information Systems Journal,
of sample size in qualitative IS studies. We intended to develop (14:4), 2004, 363-386.
evidence-based guidelines for estimating sample size. However, [15] Guest, G., Bunce, A., & Johnson, L. “How many interviews
the vast range of sample sizes for all research designs makes are enough? An experiment with data saturation and
this problematic, which further supports that following rigorous variability,” Field Methods, (18), 2006, 59-82.
methods of justifying sample size of interviews in qualitative [16] Havelka, D., & Merhout, J. W. “Toward a theory of
studies is needed. Indeed, for qualitative IS research to gain wider information technology professional competence,” Journal
acceptance relative to quantitative research, rigor in sample size of Computer Information Systems, (50:2), 2009, 106-116.
determination is critical. [17] Huang, E. Y., & Lin, S. W. “Do knowledge workers use e-
Based on our examination of qualitative interviews in IS mail wisely,” Journal of Computer Information Systems,
studies, we make the following recommendations: (a) grounded (50:1), 2009, 65-73.
theory qualitative studies should generally include between 20 [18] Kaplan, B., & Duchon, D. “Combining qualitative and
and 30 interviews; (b) single case studies should generally contain quantitative methods in information systems research: A
15 to 30 interviews; (c) qualitative researchers should examine case study,” MIS Quarterly, (12:4), 1988, 571-586.
the expectations of their intended journal outlets based on history [19] Kaplan, B., & Maxwell, J. A. “Qualitative research
and culture; and (d) replication studies should further examine the methods for evaluating computer information systems,”
impacts of culture and study design. In J. G. Anderson & C. E. Aydin (Eds.), Evaluating the
organizational impact of healthcare information systems
REFERENCES (2nd ed.), Springer, New York, NY, 2006, 30-55.
[20] Kuzel, A. J. “Sampling in Qualitative Inquiry,” In BF
[1] Amin M., & Mabe M. “Impact factors: Use and abuse,” Crabtrree and WL Miller (Eds.) Doing Qualitative Research
Perspectives in Publishing, (1), 2000, 1-6. (2nd ed.), Sage, Thousand Oaks, CA, 1999, 33-45.
[2] Bowen, G. A. “Naturalistic inquiry and the saturation [21] Lincoln, Y. S., & Guba E. G. Naturalistic Inquiry, Sage,
concept: A research note,” Qualitative Research, (8:1), 2008, Beverly Hills, CA, 1985.
137-152. [22] Miles M. B., & Huberman A. M. Qualitative Data Analysis
[3] Chan S. C. H., & Ngai E. W. T. “A qualitative study of (2nd ed.), Sage, Thousand Oaks, CA, 1994.
information technology adoption: How ten organizations [23] Morse, J. M. “The significance of saturation,” Qualitative
adopted Web-based training,” Information Systems Journal, Health Research, (5), 1995, 147-149.
(17:3), 2007, 289-315. [24] Morse, J. M. “Determining sample size,” Qualitative Health
[4] Charmaz, K. “Grounded theory for the 21st century: Research, (10:1), 2000, 3-5.
Applications for advancing social justice studies,” In N. [25] Morse, J. M. “Myth #53: Qualitative research is cheap,”
K. Denzin & Y. S. Lincoln (Eds.), The Sage Handbook of Qualitative Health Research, (12:10), 2002, 1307-1308.
Qualitative Research 3rd edition, Sage, Thousand Oaks, CA, [26] Myers, M. D. “Qualitative research in information systems,”
2005, 507-535. MIS Quarterly, (21:2), 1997, 241-242.
[5] Cheek, J. “The practice and politics of funded qualitative [27] Niederman, F., Beise, C., & Beranek, P. “Issues and concern
research,” In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage about computer-supported meetings: The facilitator’s
Handbook of Qualitative Research 3rd ed, Sage, Thousand perspective,” MIS Quarterly, (20:1), 1996, 1-22.
Oaks, CA, 2005, 387-409. [28] Nissen, M. E. “Dynamic knowledge patterns to inform
[6] Creswell, J. W. Qualitative inquiry & research design: design: A field study of knowledge stocks and flows in an
Choosing among five approaches (2nd ed.). Sage, Thousand extreme organization,” Journal of Management Information
Oaks, CA, 2007. Systems, (22:3), 2005, 225-263.
[7] Denzin, N. K., & Lincoln Y. S. Handbook Of Qualitative [29] Onwuegbuzie, A. J., & Leech, N. L. “A call for qualitative
Research. Sage, Thousand Oaks, CA, 1994. power analyses,” Quality & Quantity, (41), 2007, 105-121.
[8] Denzin, N. K., & Lincoln Y. S. “The discipline and practice of [30] Parahoo, K. “Square pegs in round holes: Reviewing
qualitative research,” In N. K. Denzin & Y. S. Lincoln (Eds.), qualitative research proposals,” Journal of Clinical Nursing,
The Sage Handbook of Qualitative Research (3rd ed.), Sage, (12), 2003, 155-157.
Thousand Oaks, CA, 2005, 1-32. [31] Patton, M. Q. Qualitative Research & Evaluation Methods,
[9] Dubé, L., & Paré, E. “Rigor in information systems Sage, Thousand Oaks, CA, 2002.

Fall 2013 Journal of Computer Information Systems 21


[32] Richardson, L., & St. Pierre, E. A. “Writing: A method of IT business value,” Journal of Management Information
inquiry,” In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage Systems, (24:1), 2007, 13-54.
Handbook of Qualitative Research (3rd ed.), Sage, Thousand [38] Tan, W., Cater-Steel, A., & Toleman, M. “Implementing
Oaks, CA, 2005, 959-978. IT service management: A case study focusing on critical
[33] Ryan, S. D., & Harrison, D. A. “Considering social subsystem success factors,” Journal of Computer Information Systems,
costs and benefits in information technology investment (50:2), 2009, 1-12.
decisions: A view from the field on anticipated payoffs,” [39] Thomson, S. B. “Grounded theory — sample size and
Journal of Management Information Systems, (16:4), 2000, validity, “ Retrieved January 29, 2010, at http://www.buseco.
11-40. monash.edu.au/research/studentdocs/mgt.pdf, 2002.
[34] Sandelowski, M. “Sample size in qualitative research,” [40] Walsham, G. “Interpretive case studies in IS research: Nature
Research in Nursing and Health, (18), 1995, 179-183. and method,” European Journal of Information Systems, (4),
[35] Strauss, A., & Corbin, J. Basics of Qualitative Research: 1995, 74-81.
Grounded Theory Procedures and Techniques, Sage, [41] Watson-Manheim, M. B., & Bélanger, F. “Communication
Thousand Oaks, CA, 1990. media repertoires: Dealing with the multiplicity of media
[36] Subramanian, G. H., & Peslak, A. R. “User perception choices,” MIS Quarterly, (31:2), 2007, 267-293.
differences in enterprise resource planning implementations,” [42] Yin, R. Case Study Research: Design and Methods (2nd ed.),
Journal of Computer Information Systems, (50:3), 2010, Sage, Beverly Hills, CA, 1994.
130-138. [43] Yin, R. Case Study Research: Design and Methods (4th ed.).
[37] Tallon, P. P., & Kraemer K. L. “Fact or fiction? A sensemaking Sage, Thousand Oaks, CA, 2009.
perspective on the reality behind executives’ perceptions of

22 Journal of Computer Information Systems Fall 2013

View publication stats

You might also like