Volume 18, No. 1, Art. 14
January 2017
Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
Brian Kelleher Sohn
Key words:
Abstract: An oft-cited phenomenological methodologist, Max VAN MANEN (2014), claims that
phenomenology;
qualitative data analysis software (QDAS) is not an appropriate tool for phenomenological research.
qualitative data
Yet phenomenologists rarely describe how phenomenology is to be done: pencil, paper, computer?
analysis software;
DAVIDSON and DI GREGORIO (2011) urge QDAS contrarians such as VAN MANEN to get over
qualitative
their methodological loyalties and join the digital world, claiming that all qualitative researchers,
research
whatever their methodology, perform processes aided by QDAS: disaggregation and
recontextualization of texts. Other phenomenologists exemplify DAVIDSON and DI GREGORIO's
observation that arguments against QDAS often identify problems more closely related to the
researchers than QDAS. But the concerns about technology of McLUHAN (2003 [1964]),
HEIDEGGER (2008 [1977]), and FLUSSER (2013) cannot be ignored. In this conceptual article I
answer the questions of phenomenologists and the call of QDAS methodologists to describe how I
used QDAS to carry out a phenomenological study in order to guide others who choose to reconcile
the use of software to assist their research.
Table of Contents
1. Introduction
2. Phenomenologists' Objections
3. General Issues and Trends with QDAS Use
4. The Technologists' Call
5. The Reconciliation
6. Recommendations
6.1 Keep your feet inside and outside the study
6.2 Bracketing
7. Conclusion
References
Author
Citation
1. Introduction
With the growing use among researchers of qualitative data analysis software
(QDAS), careful consideration must be given to its employment. It is important for
researchers, grantors, and the public to know that the use of QDAS is not a
methodology nor a marker of quality: if cabinetmakers say they use dove joints, it
indicates nothing about the value of their products. Calls for clarity regarding how
QDAS is used have been issued by some (e.g., PAULUS, WOODS, ATKINS &
MACKLIN, 2015; WOODS, PAULUS, ATKINS & MACKLIN, 2015), but examples
of QDAS use are particularly important for phenomenologists because prominent
phenomenological methodologists argue against its use. [1]
This work is licensed under a Creative Commons Attribution 4.0 International License.
Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (ISSN 1438-5627)
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
Most phenomenologists suggest reading and re-reading texts and put forth
careful explanations of the attitudinal stance that should guide ways of thinking
phenomenologically (e.g., GIORGI, 1997, 2007; SOHN, THOMAS, GREENBERG
& POLLIO, manuscript in review; THOMAS & POLLIO, 2002; VAN MANEN,
2014) or talk extensively of writing as method (e.g., VAN MANEN, 2014), but do
not sufficiently explain the tools to be used: paper, pencil, computer? A recent
review of literature regarding phenomenological methodology completely ignores
the topic of QDAS (FINLAY, 2012). [2]
One of the foremost living scholars in phenomenological methodology, Max VAN
MANEN, claims that qualitative data analysis software (QDAS) is not an
appropriate tool for phenomenological research. For VAN MANEN (2014), using
"special software" may facilitate thematic analysis in such genres as grounded
theory or ethnography, "but these are not the ways of doing phenomenology"
(p.319). He goes on to say that coding, abstracting, and generalization cannot
produce "phenomenological insights" (ibid.). [3]
Critiques of phenomenology are widespread (LANGDRIDGE, 2008; PALEY,
2017; POTTER & HEPBURN, 2005), and as a major genre of qualitative research
it is difficult to pin down, but for the sake of this article, I present a brief set of
goals for phenomenologists (see FINLAY, 2012, for a relatively thorough review
of issues in phenomenological research). Phenomenologists seek to describe or
interpret the essence and meaning of lived human experience. They seek what is
hidden beneath the accumulation of taken-for-granted assumptions that make up
most human knowledge of the world. They tend to conduct analysis without a
priori theories such as those of the social sciences, hard sciences, and poststructuralisms, with a focus instead on the every-day, first-person accounts of
phenomena. With VAN MANEN in particular, the end result of phenomenological
research must be writings that help readers "gnostically" and "pathically" to know
and feel the phenomenon in question (2014, p.268). [4]
DAVIDSON and DI GREGORIO (2011) urge QDAS contrarians such as VAN
MANEN to get over their methodological loyalties and join the digital world,
claiming that all qualitative researchers, whatever their methodology, perform the
same processes of disaggregation and recontextualization of texts—processes
aided by QDAS. They note that "arguments ... against the use of technology...are
arguments about issues that are in the realm or control of the researcher and are
not a function of the technology itself" (DAVIDSON & DI GREGORIO, 2011,
p.638). I agree that qualitative researchers are agentic actors in the choices they
make to use QDAS, notecards and highlighters, or some combination of software
and (more) manual methods. Yet we must not treat QDAS as neutral—as with
any other technology, its architecture affects its users and the research they
produce. [5]
The purpose of this article is to first examine the objections phenomenologists
have in relation to QDAS (Section 2), to then contextualize those concerns within
the growing body of literature on its use (Section 3), and then to listen to and
challenge the siren call of the technologists (Section 4). In section 5, I describe
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
the ways in which I used a particular QDAS, MAXQDA, for my dissertation, a
phenomenological case study of the student experience of other students
(SOHN, 2016). In the end I present a set of recommendations, devised through
my own work as a phenomenologist who has carefully reconciled the use of
QDAS (Section 6). [6]
2. Phenomenologists' Objections
GOBLE, AUSTIN, LARSEN, KREITZER, and BRINTNELL, phenomenologists in
the healthcare field, begin their (2012) essay with sound warnings against QDAS
framed with McLUHAN's (2003 [1964]) "medium is the message" (p.23) and
HEIDEGGER's (2008 [1977]) views on technology as dehumanizing. They
describe the ways QDAS affected them and their work, but in the end they make
precisely the kind of arguments DAVIDSON and DI GREGORIO (2011, see
above) disparage. GOBLE et al. (2012) argue that "through our use of technology
we become functions of it" (§1). It is difficult to avoid becoming a tool of our tools:
to someone with a hammer, everything looks like a nail. But hammers are the
least of our technology—mobile devices provide a more relevant example of how
we are changed by the technologies we use. FLUSSER (2013) writes that
technology is a trick that extends our reach. Using the lever as an example, he
describes the allure of cheating death through leveraging the power of art,
artifice, and machines. The reach new technologies provide is a difficult potential
to ignore, and for GOBLE et al. (2012), they conducted many more interviews
(53) than they normally would have, influenced by the presumed facilitative power
of QDAS. [7]
The right tool for the job makes it "easy." It is the facilitation provided by QDAS
that GOBLE et al. (2012) take issue with. As ADAMS (2006) notes in regard to
PowerPoint software, the majority of users employ default templates, limiting their
thinking in particular ways and rendering their presentations sequential and
numbing. The processes QDAS facilitates are, for GOBLE et al. (2012), obstacles
to phenomenological insight. They describe the problems QDAS brings to their
research and how it affects their being in the world. [8]
The first problem GOBLE et al. discuss is coding. As adherents of VAN MANEN's
phenomenology, they note that coding is unnecessary. Instead they engage in
three levels of reading: "wholistic," "selective," and "for detail" (VAN MANEN,
2014, p.320). The goal of the reading is to become familiar with the texts. While
VAN MANEN does recommend highlighting and note-taking while reading, he
does not call it coding. But when using QDAS, GOBLE et al. (2012) code so
much—because it is easy, because they have so many transcripts, and because
they are working with a large research team—that they come to see what they
have done as coding for coding's sake. They argue that QDAS leads them to see
their interview texts as data, zeros and ones that were labeled and organized in a
fashion that numbs them (ADAMS, 2006) and prevents them from being or
dwelling with the interview texts in the way required to discover their essence. [9]
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
The inability to be with and resonate with the texts is antithetical to the spirit of
phenomenology and leads to another problem: the divide that forms between the
researchers and their data. GOBLE et al. say that when using QDAS they
"become separate and distinct from [their] research" (2012, §41). The immersion
typically associated with phenomenological research eludes them because they
fret over questions such as, "can this be coded?" or "can we enter that into
NVivo?" The locale of the research and its analysis, due to the distance they feel,
prevents them from the kinds of connection they typically experience, so they
employ manual methods to overcome this issue. [10]
GOBLE et al. state that the work of phenomenology is inevitably transformed for
the worse with the use of QDAS. They conclude with platitudes such as "with
[QDAS] we become certain" (§43) and "with [QDAS] we become language-less"
(§44). These poorly-defined problems are not inherent to the technology. If these
researchers are swept away while using QDAS, if they make no attempt to
familiarize themselves with the common problems researchers have identified
with QDAS (e.g., GILBERT, 2002), they must bear at least some of the blame.
Had these researchers taken on the large research project they described and
avoided the use of QDAS, there is no indication they would have been more
successful. [11]
GOBLE et al. (2012) conclude with a dire set of proclamations: that they become
"standing reserves" with no potential for messiness and that research becomes
just "a problem to be identified and eventually solved" (§44-45). Their
underestimation of human ability to make a mess out of anything ordered is
noteworthy, as is their "certain[ty]." But whether or not the utilization of QDAS
takes users to the depth of dehumanization described by GOBLE et al., there is a
set of issues worthy of examination by the qualitative research community
generally and phenomenologists in particular: WOODS et al. (2015) found that
phenomenologists are not using QDAS as much as other researchers such as
grounded theorists (their review was limited to the use of the ATLAS.ti and NVivo
QDAS platforms). So whether or not GOBLE et al. (2012) represent an
exaggerated experience is not as important as carefully examining issues of
QDAS use. [12]
GOBLE et al. (2012) admit that they as the primary researchers were novice
QDAS users and for some this may invalidate their critiques. But one does not
have to be an expert user to critique a program. Expertise can be the source of
blind spots, particularly if one develops automaticity and fails to note how QDAS
may affect research processes; it could become a kind of invisible partner.
Secondly, they note the issue of placing qualitative research into the rat race of
corporatized university systems where they, and other researchers, feel the
pressure to use QDAS, whether or not they have experience doing so (see
MORSE, 2006 for a discussion of some of the issues leading qualitative
researchers in that direction). GOBLE et al. (2012), following VAN MANEN, want
to present the research phenomenon through "texts [that] speak to readers at an
intuitive as well as cognitive level, creating a way of understanding that is
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
embodied" (§15). How can one produce a text that is to be embodied when
he/she feels forced to use (presumably) disembodied, computerized methods? [13]
The issues phenomenologists face with QDAS are notable, and it is conceivable
that the habits of mind (ADAMS, 2006) encouraged by QDAS could impede the
processes required to do high quality phenomenological (or other qualitative)
research. Some of the specific issues with QDAS that GOBLE et al. (2012)
discuss are common among qualitative researchers, and I describe them in a
larger context in the next section. [14]
3. General Issues and Trends with QDAS Use
Novice users of technologies tend to employ the default functions of a design
(ADAMS, 2006) and the consequences of such use within QDAS have been
theorized (FRIESE, 2011) and documented (GILBERT, 2002). For example,
multiple empirical studies on the use of QDAS have found that many users spend
too much time coding (GARCÍA‐HORTA & GUERRA‐RAMOS, 2009; GILBERT,
2002), as did GOBLE et al. (2012). As FRIESE (2011) notes, before the advent of
QDAS, "[n]o one would ever come close to 1000 or more codes when using the
old-style paper & pencil technique" (§12). Identifying the problem is not the same
thing as its solution, but it can begin a discussion. [15]
Commonly referred to as the "coding trap" (GILBERT, 2002, p.218), researchers
fall into some combination of what they have learned qualitative research involves
(coding) and what the program facilitates. MAXQDA, for example, allows for a
drag and drop from codes to data or from data to codes. A click of the mouse and
a few taps on the keyboard and a new code can be created. There is a perceived
quickness and neatness afforded by such a process as compared to shuffling
notecards and writing in pencil. GARCÍA-HORTA and GUERRA-RAMOS (2009)
refer to the potential to get sucked in to various features of the technology as
"data fetishism" and note that coding everything can "[inflate] the results that are
to be reported" (p.163). [16]
From a phenomenological standpoint, the facility of coding has the potential to
instrumentalize a process that is supposed to be artful, intuitive, and lead to
"creative leaps" (CROSS, 2011, p.127) and the use of abductive reasoning (VAN
MANEN, 2014). The purpose of phenomenology as VAN MANEN states it is not
aided by piles of codes that support a certain theme. Quantity does not provide
quality, and phenomenology is aided by uniqueness, not repetition. But repetition
is easy with QDAS: GILBERT (2002) quotes one researcher who got into a kind
of coding "zone" and "lost sight of where [he was] going, what [he was] analyzing"
(p.219). [17]
In FRIESE's (2011) "Computer-assisted [Noticing, Collecting, and Thinking] NCT
analysis" (§7), collection is central to her process and if there are not a sufficient
number of collected coded segments in a category, it "create[s] analytic
problems" (§20). It seems that grounded-theory-influenced researchers such as
FRIESE suggest the adoption of a tendency of quantitative research: to ignore
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
outliers. It is possible that analytic categories with few codes represent a critical
aspect of a research project. On the other hand, perhaps the category deserves
to be ignored, but when working with QDAS, the potential to lose a crucial aspect
of a phenomenon can be high for no other reason than it lacks quantity. [18]
Getting bogged down in quantities and the nitty gritty details of coding can take
unnecessary time and effort, and can prevent the researcher from gaining a
sense of the whole of a phenomenon. For phenomenologists, getting caught up
in coding would be most beneficial if, through dwelling and getting lost in the
words of participants, they find an example that "reconciles the incommensurable
couplet of the particular and the universal" (VAN MANEN, 2014, p.260). But if
coding were in some way mindless, it would be a waste of time. It is how one
approaches the QDAS and coding that can make this difference. [19]
The coding trap is partially a trap because, as GILBERT (2002) points out,
analytical power is reduced if the researcher's focus is zoomed in completely on
phrases. Distance from the data is required to see patterns or intuit an overall
sense of an interview transcript. To know whether or not phrases from various
participants that are not literally the same represent an "experiential pattern"
(THOMAS & POLLIO, 2002) requires the ability to step away from the particular
—the gestalt of the individual interview, for the moment, must be broken. Yet if
the researcher feels a coalescence across those dissimilar, particular phrases, a
sameness can be asserted. The echo in the other interviews can enrich the
particular gestalt of a single individual interview. [20]
This raises an interesting point for GOBLE et al. (2012), who claim that distance
is a problem in their use of QDAS. Generally they object that they cannot
immerse themselves in the data because of the barriers imposed by the
technology. But many QDAS users feel too close to their data (GILBERT, 2002),
and if GOBLE et al. fell into the coding trap, it is interesting that they coded for
coding's sake while feeling distant. Overuse of technical features such as the
ability to highlight, click, and code segments of text, rather than draw them in,
raised a barrier. Their goal, to find the essence of a phenomenon, was apparently
impeded though the segmenting of interview transcripts. This difficulty, even if
separated from the use of QDAS, must be considered when authors like
DAVIDSON and DI GREGORIO (2011) say all qualitative researchers essentially
do the same things. [21]
4. The Technologists' Call
When I heard of QDAS for the first time I was with a doctoral student who
showed me the basic features of ATLAS.ti. I messed around with coding and
memoing and in subsequent years continued to develop the impression that
using QDAS was more efficient, more legitimate, and the thing to do: our
university provided free access to three QDAS platforms to encourage their use. I
found more direct calls to use QDAS in the literature. [22]
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
DAVIDSON and DI GREGORIO (2011), some of the top scholars of QDAS,
describe the basic functions of QDAS as disaggregation and recontextualization
of the data (p.633). They argue that these functions can serve any and all
qualitative research paradigms and that any proclaimed differences between the
genres are "residue" (p.639) from battles over legitimacy. They urge all qualitative
researchers to get over their sad clinging to tradition and join the digital
revolution. [23]
This gloss of qualitative research fails to account for substantial differences in
grounded theory, ethnography, narrative analysis, and phenomenology, and is
typical language harkening to general technology adoption processes. "Come on,
luddites, you're slowing us all down." The differences between the products of
various qualitative research genres are stark depending on the tools used.
Ignoring the differences is only possible when taking a broad view of qualitative
research in which all researchers embark on "an iterative process of identifying
the questions to be addressed, using the tool to access the data that could
illuminate those questions, and through a process of exploration, retrieval, and
comparison develop the analysis" (DI GREGORIO, 2011, §8). The same
description could be used to define quantitative research or problem solving
generally. When you examine an article like Catherine ADAMS' (2006)
phenomenological study of PowerPoint, you can see that phenomenology (when
true to a specific brand of phenomenology, in her case VAN MANEN's
hermeneutic phenomenology), would not have clearly been served by
disaggregation and recontextualization of her "data," if she would even have
referred to the texts she calls on with that term. [24]
DAVIDSON and DI GREGORIO (2011) cite the work of LEWINS and SILVER
(2007) to support the idea that qualitative researchers all engage in similar
activities, but POTTER (1996), among the first researchers to conduct an
empirical meta-analysis of qualitative methods, created a typology that illustrated
clear differences in axiology, process, and final product. Each research genre has
particular epistemologies and ontologies. However, other researchers have noted
the similarities across published qualitative research studies and note the trouble
defining the distinctiveness of different approaches in practice (PALEY, 2017). If
researchers are not familiar with the paradigms of their genre before they begin
using QDAS, they are likely to use default functions of QDAS. This can lead to a
unification of different qualitative research methods and reduce the complexity of
epistemologies and ontologies and therefore the possibility for rare and unique
insights that can be gleaned with such different approaches as discursive
psychology (EDWARDS & POTTER, 1992) and phenomenology, for example. [25]
Such "qualification" of the different qualitative research genres has been
documented in a review of literature by WOODS et al. (2015). Focusing on
ATLAS.ti and NVivo, they found that 326 of the 763 studies only went as far as
describing their methods as "qualitative" or as "interview studies." These generic
studies are at greater risk of being interpreted as a "just so" account of
phenomena and obfuscate the differences between the maps and the territory. I
do not discount the need for broad views of social problems, but if qualitative
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
research is reduced to inductive reasoning and post-positivistic interpretation, we
lose the beauty that can be found in the subtleties and exceptions so essential to
learning, social justice, and advocacy. [26]
Of the studies in WOODS et al. review, 31 out of 763 employed phenomenology,
as compared to 100 which were grounded theory. Even with its well-known
history and popularity, grounded theory was still far less common than the
generic genres of "qualitative" or "interview study." This dominance may occur
due to funding agencies and priorities of journal editors, but another possibility is
the influence of QDAS. If one takes seriously DAVIDSON and DI GREGORIO's
(2011) idea that all qualitative researchers engage in similar activities, and one
sees that many qualitative researchers are shirking the labels of traditional
approaches, whether or not researchers claimed a particular research genre
becomes less interesting than what procedures they used. [27]
WOODS et al. (2015) found that what researchers usually do with QDAS is
"support coding and retrieval of data, differentiate coded data by participant
characteristics, and investigate conceptual relationships ... [they also] make
analytic processes more transparent, primarily by using program outputs to
illustrate their coding processes and research outputs" (p.14). Their conclusion is
not dissimilar from DI GREGORIO's (2011) commentary on an experimental
conference in which researchers were asked to analyze the same data set with
the same research questions using different QDAS platforms. But use, strictly
speaking, is not the primary concern of the phenomenologist—rather the
important question posed by GOBLE et al. (2012) is, how does using these
features affect the researcher's being in the world? The processes DI
GREGORIO (2011) and WOODS et al. (2015) described as common among
QDAS users are the same ones VAN MANEN (2014) critiques (save for the idea
that the analysis process can be more transparent with QDAS—one has a kind of
paper trail [KONOPÁSEK, 2008]). But do they preclude a researcher from gaining
phenomenological insight? And are phenomenologists using QDAS differently?
There is a difference between writing with a pencil and typing with a computer,
clicking and dragging segments of text and creating a notecard. In the next
section I describe how I used QDAS—wary of its affects but eager to take
advantage of its conveniences. [28]
5. The Reconciliation
QDAS is often described as a tool (DI GREGORIO, 2011; GILBERT, 2002), but it
could also be referred to as a structure. A structure provides support, but also
imposes limits. The particular supports or limitations of QDAS can affect the
being-in-the-world of the researcher. In my dissertation study (SOHN, 2016),
which focused on the student experience of other students in a graduate seminar,
I used MAXQDA to organize my documents, identify, label, index, and retrieve
text, and to review the development of conceptual understanding of my data. As a
phenomenologist in the field of education, I was guided by the writings of
THOMAS and POLLIO (2002) and VAN MANEN (2014). I was supported (and
limited) by the use of MAXQDA during the analysis and the writing phases of my
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
dissertation. The goal of this section is not to determine with certainty whether or
not MAXQDA was boon or burden but to describe my experience of its use. [29]
I chose to use MAXQDA because I knew my dissertation project was much
bigger than any project I had undertaken previously and my paper organization
skills are poor. MAXQDA was the only QDAS I could purchase and keep—other
programs were sold with yearly licenses. I had no preferences and minimal
experience using any QDAS platform—I had recently used MAXQDA to analyze
30 interviews for another research project (PETTIGREW, SOHN, DALTON,
CASTILLO & ALLSUP, manuscript in review). [30]
I did not use MAXQDA during the preparation phases of my study. I developed
my question, reviewed literature, refined my question, and engaged in a
bracketing interview outside of any QDAS platform. I analyzed my bracketing
interview with an interdisciplinary phenomenology research group (IPRG) (SOHN
et al., manuscript in review; see also THOMAS & POLLIO, 2002) that assisted
me in developing an awareness of my positionality regarding my study. I kept a
research journal using a word processor, and later copied and pasted it into
MAXQDA's logbook feature. I experienced this transfer with a feeling of now I'm
really beginning and a mixture of anxiety and anticipation. [31]
I began to use MAXQDA by uploading interview transcripts and other documents.
Unlike some QDAS users, I had engaged in manual coding of some of the printed
transcripts before using MAXQDA. Much of this manual coding took place during
meetings of the IPRG. My project was a new analysis of data I had been working
with for four years as part of a research team investigating phenomenological
teaching and learning (e.g., GREENBERG, GREENBERG, PATTERSON &
POLLIO, 2015; SOHN et al., 2016). This familiarity may have helped me avoid
thinking of the documents as static collections of words to be clicked on for data
access—I had heard many of their passages read aloud and discussed in our
research group meetings. [32]
I began with in vivo coding to identify meaning units in the documents (SOHN,
2016; THOMAS & POLLIO, 2002). I cannot say definitively I managed to avoid
the coding trap. In my 25 main documents, I coded 2,290 segments. There was
one group of codes that did not relate directly to the findings I eventually
reported, and another group of codes used to make re-organized documents for
further analysis. Was the ease of coding what led me to develop those unused
sets of codes? I certainly would not have developed the codes designed to create
new documents without a QDAS, although I would have spent a lot of time
copying and pasting in a word processor. Would I have immediately coded more
useful segments had I employed strictly manual methods? In my relatively brief
experience doing qualitative research (five years), I know that I always have
some false starts, with or without a software platform architecture to influence my
analysis. In this project, I felt that a fumbling towards what I needed in terms of
coding was to be expected. Even though at the time I regretted many of the
codes I had created, in the long run I felt that part of my facility in recalling
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
random segments of data was at least partially attributable to the totality of my
coding process. [33]
A potential problem with coding in phenomenology is to lose the context of the
coded segment of text, but with a click, MAXQDA retrieves the context of a coded
segment (see Figure 1). Whether or not the coding I did was always useful to the
final report, it was always a way to further immerse myself in the data, and none
of my coding or memoing made me feel that I was dealing with anything other
than the words of a living, breathing human. Many phenomenological
methodologists call for line-by-line analysis. QDAS facilitated this, and I did not
lose the sense of discovery, exploration, and wonder that is supposed to drive
phenomenological inquiry. As a life-long rock climber, I have established routes in
unexplored areas—I know a sense of freedom when I feel it. And QDAS did not
limit this feeling as I coded, re-coded, and wrote analytic memos and research
journal entries in MAXQDA. Even if one followed VAN MANEN's dictate that
coding is not part of phenomenology, one could read "wholistically ...
selectively ... and for detail" (2014, p.320) and write analytic memos of thoughts
for each word, line, paragraph, and document, as he recommends.
Figure 1: Retrieval and context in MAXQDA. In the right-hand column are codes retrieved
by selecting documents and codes (highlighted in the left column). By clicking on the
retrieved segments in the right-hand column, their context appears in the middle column.
Please click here for an increased version of Figure 1. [34]
When GOBLE et al. (2012) entered their thoughts into QDAS memos, they had a
sense that the thoughts were minimized. But memos are markers that can bring
the researcher back to the entire cognitive and non-cognitive experience of a
reaction to research data. I used memos relatively frequently: I created 193 total,
and they often served to re-orient me to times when a line of data provided
revelation. Re-reading my memos from the previous day of work would help me,
a father of a toddler at the time of the research project, jump back in to the
analysis. Unlike GOBLE et al., I was not less immersed in the data because of
QDAS; rather the virtual setting of the texts seems to have assisted me in
stepping out of my everyday life and into the world of the study. I could hear the
voices of participants by listening to recorded audio from the class sessions or
interviews, read their reflections, and step into their experiences alongside them.
Having the data and codes and memos all in one virtual place allowed me
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
distance from the data as well, a key to the abductive thinking required for
phenomenological insight. [35]
Every four to six weeks while I was conducting my analysis in MAXQDA, I took
printed transcripts to an IPRG for discussion, insight, confirmation, contestation,
and continued bracketing (SOHN et al., manuscript in review). In these sessions
other members would read transcript sections aloud and describe what stood out
to them about the student experience of other students. These sessions often
sparked ideas for further review of literature. For example, in reading a transcript
in which a study participant complained about group work, a veteran member of
the group recommended I read a prior study that included the famous line from
SARTRE's (1989 [1947]) play "No Exit," "Hell is other people" (p.46). The
discussion continued with references to more relational phenomenologists such
as MERLEAU-PONTY (e.g., 1962 [1945]) and BUBER (e.g., 1970 [1937]). These
references, which were called to mind by the transcripts, enhanced my
understanding of the human condition in relation to other people, the central
concern of my study. [36]
Such departures during the analysis phase were indispensible interruptions from
the work of reading and re-reading the data and writing about it within MAXQDA.
The importance of closeness to and distance from data in qualitative research
has been well discussed (e.g., GILBERT, 2002): we need to be immersed in our
data and zoom in on the particular, but we also need to use a wide angle view in
order to successfully engage in interpretation or analogy. The research group
provided the opportunity to dialogue with veteran and novice phenomenologists
with positionalities distinct from my own—they focused in on details I may have
skipped and shared broad views on the data that often aligned with mine but
sometimes did not. [37]
When I reached what I felt was a powerful understanding of my data (a position
that developed over nine months), I began writing the results and discussion
sections of my dissertation. It is here, I thought, that all the hard work of coding,
indexing, and memoing would pay off. I could just copy and paste my results
chapter into existence! And this was the most surprising drawback of using QDAS
—the ease with which I could retrieve segments of texts turned writing into a
chore. I began by copying and pasting into an outline, but after a few lackluster
days I found myself avoiding writing despite looming deadlines. [38]
For me, it was this step—going from a finished analysis in MAXQDA to the writing
of the report—where phenomenology suffered the most. I can imagine that for a
grounded theory or content analysis study, copying and pasting from the retrieve
functions of a QDAS would be time-saving. But copying and pasting does not serve
when the goal of phenomenological writing is, as MERLEAU-PONTY puts it, to
"[bring] the meaning [of a phenomenon] into existence as a thing as the very heart of
the text, it brings it to life in an organism of words, establishing it in the writer or the
reader as a new sense organ, opening a new field or a new dimension to our
experience" (MERLEAU-PONTY, 1968 [1964], p.182). [39]
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
To break out of the MAXQDA-facilitated writing rut, I returned to reading
MERLEAU-PONTY and VAN MANEN (2014). Van MANEN writes that the goal of
phenomenological writing is to "bring experience vividly into presence" (p.241).
The writings of these phenomenologists helped me to re-examine my data
interpretations to find what in them stirred me. I prioritized those findings and got
back to work with MAXQDA as an assistant to, rather than the driver of my
writing. [40]
6. Recommendations
QDAS is not a neutral tool, but we as researchers can develop deep foundations
in the epistemologies and ontologies of our genre to avoid the traps some
methodologists claim are inherent in its use. As a researcher new to using QDAS,
SCHUHMANN (2011) worried that the structure of QDAS platforms would limit
creative interpretation. She later concluded that the interface with a QDAS
platform "adds a layer of interpretation to qualitative analysis as one has to know
how to 'read' a software package" (§2). This additional layer, the interface
between user and QDAS platform, is where the following recommendations will
best serve researchers. In my case, I found that I coded to immerse myself in my
data without feeling as if it were chopped into little bits. I analyzed within
MAXQDA using the memo and logbook features without over-systemizing my
thoughts. When I did feel as if MAXQDA was hurting my study, I returned to the
IPRG and reread phenomenological writings to reignite my motivation for
producing the report. [41]
Discovery, openness, and "wonder in the face of the world" (MERLEAU-PONTY,
1962 [1945], p.xiii), are some of the most common features of the
phenomenological attitude. In order to maintain these qualities while using QDAS,
I recommend the following practices: keep your feet inside and outside the study
and be diligent and exhaustive in bracketing. [42]
6.1 Keep your feet inside and outside the study
The question of closeness and distance is common in qualitative research (e.g.,
GILBERT, 2002), but these discussions are often limited to the arena of data. To
reconcile phenomenology and QDAS, we must extend the discussion to other
areas. I include the following: stay steeped in the epistemologies and ontologies
of the research genre, use manual and computerized methods to be intimate with
the data, and interrupt analysis. [43]
To be immersed in phenomenology, throughout my study I read the works of wellknown phenomenologists and phenomenological research reports and discussed
them with members of the IPRG. For those without a research group, a colleague
with similar interests may suffice. We know that particular theoretical lenses
provide unique insights—just as Black studies reveal universal truths about social
injustice, empirical studies following a series of procedures clearly and logically
influenced by philosophy are likewise uniquely suited to provide novel ideas
regarding research phenomena. [44]
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
The potential for phenomenology to lose its distinctiveness can be countered by
staying aware of the directions QDAS can lead. If GOBLE et al. (2012) fell asleep
at the wheel of QDAS, other phenomenologists can remain awake by maintaining
an intimate presence to their data inside and outside the world of the software.
While it may be impractical for research assistants, the time I spent journaling,
listening to audio recordings, and auditing transcripts was helpful in keeping a
sense of the humanity and wholeness of the various forms of data collected for
my study. With all the apps and add-ons available for the major QDAS platforms
available now, one need never conduct any part of a qualitative research project
with manual methods, but for phenomenologists, doing so will help develop a
sense of the essence of the research phenomenon that can grow and become
more clearly differentiated in subsequent steps of the analysis. [45]
If data begins to feel stale, if dehumanization is a problem, find ways to re-enliven
research texts. Re-listen to audio recordings of interviews. The humanity,
emotion, and pathic elements often missing from a researcher's overly cognitive
focus while reading texts on a screen can be re-lived intimately through hearing
the voices of study participants. It is during these times that emotions may well up
within the researcher, spurring a feeling for the essence of the phenomenon.
These steps can be facilitated by QDAS platforms such as Transana (see
DEMPSTER & WOODS, 2011, §62), which directly link transcription files to the
media from which they were derived. [46]
To interrupt my analysis, I took regular breaks from my dissertation work to do
the work of academia: develop manuscripts, submit to journals and conferences,
and prepare to teach courses. As VAN MANEN (2014) notes, exercise and
similar activities allow the brain to engage in what he refers to as "active
passivity" and "passive activity" (pp.345-346). I did not always see the broad
experiential patterns I sought during the moments of active activity, or purposeful
engagement, with the data within MAXQDA. In times of passive activity or active
passivity—driving, teaching, cooking, playing with my son, jogging—I gained
insights to develop themes. [47]
6.2 Bracketing
Bracketing, a critical element of the phenomenological attitude, must be
understood and practiced with diligence when using QDAS. Bracketing must take
place within and without the QDAS platform, and I recommend approaches to
maintaining wonder, coding, and thematizing. [48]
The metaphor of the stranger (see SHABATAY, 1991) is a useful one in
negotiating bracketing when using QDAS. If we are to maintain wonder in the
face of the data on the screen, the phrases and texts in the windows, we must be
well aware of what we "know," yet take on the next participant's interview
transcript with an openness that allows for contradiction or confirmation. With
QDAS one could immediately go to a phrase called to mind in a recently reviewed
transcript rather than stay with the current one. In order to "slackens
the...threads" such that "the forms of transcendence fly up like sparks from a fire"
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
(MERLEAU-PONTY, 1962 [1945], p.xiii) (or from the computer screen), we must
use the coding and memoing features to document and set aside initial and
developing impressions that may not be resident in the transcript of another
participant. We can allow the text to stand on its own, be its own world, and enter
this world as a stranger. [49]
If we read, code, and memo in too systematic a way, a simplistic focus can
dominate our orientation to the data. Piles of codes can be collected into
categories without difficulty, potentially clouding our view. Yet at the same time
they can provide ways to gain a sense of literally and figuratively similar elements
within a data set—the phrases of various participants can maintain their
particularity even as they are connected through the intentionality of the
researcher. This is especially true when we take advantage of QDAS features
that allow us to examine coded segments in their original context. [50]
Finding literal similarity may be enhanced by QDAS, but such commonalities may
do little to enhance understanding of the essence of a phenomenon. In order to
develop broad experiential patterns as themes, in order to experience the
breakthroughs associated with the interpretation of the structure of human
experience, metaphors can be mined for meaning (THOMAS & POLLIO, 2002).
We must also experience the echo of resonance between the data and ourselves
—and taking a moment to document such feelings in a memo may diminish it
(memos, as pop-up features, are indeed smaller than other windows), but I do not
believe that writing about it with a pencil would be very different. When QDAS
mediates this experience, there is potential with coding, memo, and logbook
features to further document and interrogate not only the analytic incidents that
preceded the epiphany, but also the realization itself. Another way to document
and interrogate phenomenological insights is through a research group. [51]
Talking with other colleagues and members of my research group was one of the
ways I was able to continue bracketing and step beyond my own perceptions of
my data. Our own perspective is powerful—we are most intimate with our
research project. But colleagues can broaden the horizon upon which we see the
figural aspects of our data. Their assistance in questioning our interpretation
helps us with bracketing—both in a reassertion of the generative power and
relevance of our prior knowledge and in the illumination of what we have taken for
granted. The group can ask seemingly naïve questions that, if taken seriously,
can help researchers know what they know in a more profound way. As is the
case with many other facets of qualitative research, even discussing
interpretations with colleagues can occur within QDAS platforms (e.g.,
DEMPSTER & WOODS, 2011) if face to face meeting is difficult. However, the
commitment required of a face-to-face meeting signals the potential losses that
accompany virtual feedback. LEVINAS's (1979 [1961]) insight, that in the face of
another I see my responsibility, is surely relevant here. [52]
An early version of this article was presented at the Ethnographic and Qualitative
Research Conference in Las Vegas, NV, in 2016. The audience response was
generally positive, but one audience member asked me afterwards how many
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
pages of data were involved in my dissertation project. "Around 500," I told him.
He laughed haughtily. "With that you can keep it all in your head. Now 1000 or
more, then you need a computer." He said the project was changed for the worse
by my use of QDAS. But the study would have taken me longer without QDAS,
and its quality may have been the same. If one possesses a photographic
memory, perhaps QDAS is unnecessary. [53]
7. Conclusion
Phenomenologists and technologists tend to be in opposition. The majority of
well-known phenomenologists, save GIORGI (2007) and PERRY (e.g., 1998), are
known for their love of the humanities and share Heidegger's concern that
through technology humans may become exploitable resources. But software
developers must accept that users will typically employ the default settings of
QDAS. Researchers must have a sound basis in their epistemologies and
ontologies before learning to use QDAS. As noted by DAVIDSON and DI
GREGORIO (2011), "all tools have limitations and ... tools for research are in
constant flux and development. Therefore, researchers must engage in an active
dialogue between methodology and technology in order to craft the appropriate fit
for their work" (p.633). Researchers who use QDAS without the direction of welldeveloped ontologies and epistemologies will be directed by the architecture of
the software. If researchers have not found a methodological home, it is likely
they may end up writing a generic qualitative study that, due to lack of a distinct
perspective, fails to provide any profoundly new insights. [54]
Like artists, phenomenologists do not want to be confined. They want some of
their method to be ineffable. We cannot always explain ahead of time the best
methods to explore a phenomenon or why a text stirs us deeply. But for me (and
many other phenomenologists), QDAS is not a tyranny. QDAS does not force us
into non-phenomenological habits of mind. QDAS is not in any sense a neutral
tool—researchers must be aware not only of what QDAS can and cannot do, but
how it does and does not affect their being-in-the-world. But with a careful and
dedicated attention to phenomenological considerations such as wonder and
lived experience, phenomenologists can reconcile the use of QDAS to try and
avoid its potential pitfalls. [55]
References
Adams, Catherine (2006). PowerPoint, habits of mind, and classroom culture. Journal of
Curriculum Studies, 38(4), 389-411. http://doi.org/10.1080/00220270600579141
Buber, Martin (1970 [1937]). I and thou. New York: Charles Scribner's Sons.
Cross, Neil (2011). Design thinking: Understanding how designers think and work. Oxford: Berg
Publishers.
Davidson, Judith & di Gregorio, Silvana (2011). Qualitative research and technology: In the midst of
a revolution. In Norman Denzin & Yvonna Lincoln (Eds.), Handbook of qualitative research (4th ed.,
(pp.627-643). London: Sage.
Dempster, Paul G. & Woods, David K. (2011). The economic crisis though the eyes of Transana.
Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 12(1), Art. 16, http://nbnresolving.de/urn:nbn:de:0114-fqs1101169 [Accessed: December 22, 2016].
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
di Gregorio, Silvana (2011). Comment: KWALON Conference: Is qualitative software really
comparable? Reflections on "the experiment": An "expert" view. Forum Qualitative
Sozialforschung / Forum: Qualitative Social Research, 12(1), http://nbnresolving.de/urn:nbn:de:0114-fqs1101C35 [Accessed: December 18, 2016].
Edwards, Derek & Potter, Jonathan (1992). Discursive psychology. London: Sage.
Finlay, Linda (2012). Debating phenomenological methods. In Norm Friesen, Carina Henriksson &
Tone Saevi (Eds.), Hermeneutic phenomenology in education (pp.17-37). Rotterdam: Sense
Publishers.
Flusser, Vilém (2013). The shape of things: A philosophy of design. London: Reaktion Books.
Friese, Susanne (2011). Using ATLAS.ti for analyzing the financial crisis data. Forum Qualitative
Sozialforschung / Forum: Qualitative Social Research, 12(1), Art. 39, http://nbnresolving.de/urn:nbn:de:0114-fqs1101397. [Accessed: December 18, 2016]
García‐Horta, José & Guerra‐Ramos, María (2009). The use of QDAS in educational research:
Some advantages, limitations and potential risks. International Journal of Research & Method in
Education, 32(2), 151-165. http://doi.org/10.1080/17437270902946686
Gilbert, Linda (2002). Going the distance: "Closeness" in qualitative data analysis software.
International Journal of Social Research Methodology, 5(3), 215-228.
http://doi.org/10.1080/13645570210146276
Giorgi, Amedeo (1997). The theory, practice, and evaluation of the phenomenological method as a
qualitative research procedure. Journal of Phenomenological Psychology, 28(2), 235-260.
Giorgi, Amedeo (2007). Concerning the phenomenological methods of Husserl and Heidegger and
their application in psychology. Collection du Cirp, 1, 63-78, http://www.cirp.uqam.ca/documents
%20pdf/Collection%20vol.%201/5.Giorgi.pdf [Accessed: November 18, 2016].
Goble, Erika; Austin, Wendy; Larsen, Denise; Kreitzer, Linda & Brintnell, Sharon (2012). Habits of
mind and the split-mind effect: When computer-assisted qualitative data analysis software is used
in phenomenological research. Forum Qualitative Sozialforschung / Forum: Qualitative Social
Research, 13(2), Art. 2, http://nbn-resolving.de/urn:nbn:de:0114-fqs120227 [Accessed: November
12, 2015].
Greenberg, Neil; Greenberg, Katherine; Patterson, Russell & Pollio, Howard (2015). Patterns in
transformative pedagogy: Ethological perspectives. Poster presented at the 7th Annual Conference
on Higher Education Pedagogy, Blacksburg, Virginia Polytechnic and State University, USA,
February 5, 2016.
Heidegger, Martin (2008 [1977]). The question concerning technology. In David Farrell Krell (Ed.),
Basic writings (pp.311-341). New York: Harper Perennial.
Konopásek, Zdenek (2008). Making thinking visible with ATLAS.ti: Computer assisted qualitative
analysis as textual practices. Forum Qualitative Sozialforschung / Forum: Qualitative Social
Research, 9(2), Art. 12, http://nbn-resolving.de/urn:nbn:de:0114-fqs0802124 [Accessed: July 8,
2014]
Langdridge, Darren (2008). Phenomenology and critical social psychology: Directions and debates
in theory and research. Social and Personality Psychology Compass, 2(3), 1126-1142.
Levinas, Emmanuel (1979 [1961]). Totality and infinity: An essay on exteriority (transl. by A. Lingis).
London: Martinus Nijhoff Publishers.
Lewins, Ann & Silver, Christina (2007). Using software in qualitative research: A step-by-step guide.
Thousand Oaks, CA: Sage.
McLuhan, Marshall (2003 [1964]). Understanding media: The extensions of man. Toronto, ON:
McGraw Hill.
Merleau-Ponty, Maurice (1962 [1945]). Phenomenology of perception (transl. by C. Smith). New
York: Routledge.
Merleau-Ponty, Maurice (1968 [1964]). The visible and the invisible (ed. by C. Lefort, transl. by A.
Lingis). Evanston, IL: Northwestern University Press.
Morse, Janice (2006). The politics of evidence. Qualitative Health Research, 16(3), 395-404.
Paley, John (2017). Phenomenology as qualitative research: A critical analysis of meaning
attribution. New York: Routledge.
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
Paulus, Trena; Woods, Megan; Atkins, David & Macklin, Rob (2015). The discourse of QDAS:
Reporting practices of ATLAS.ti and NVivo users with implications for best practices. International
Journal of Social Research Methodology, 20(1), 35-47.
Perry, Beth (1998). Beliefs of eight exemplary oncology nurses related to Watson's nursing theory.
Canadian Oncology Nursing Journal/Revue canadienne de soins infirmiers en oncologie, 8(2), 97101.
Pettigrew, Jonathan; Sohn, Brian; Dalton, Betsy; Castillo, Maria & Allsup, Josh (manuscript in
review). A qualitative analysis of adolescent drug resistance in Nicaragua, Central America:
Resistance strategies and motivations for non-use. Journal of International & Intercultural
Communication.
Potter, James (1996). An analysis of thinking and research about qualitative methods. Mahwah, NJ:
Lawrence Erlbaum.
Potter, Jonathan & Hepburn, Alexa (2005). Qualitative interviews in psychology: Problems and
possibilities. Qualitative Research in Psychology, 2(4), 281-307.
http://doi.org/10.1191/1478088705qp045oa
Sartre, Jean-Paul (1989 [1947]). No exit and three other plays. New York: Vintage.
Schuhmann, Carmen (2011). Comment: Computer technology and qualitative research: A
rendezvous between exactness and ambiguity. Forum Qualitative Sozialforschung / Forum:
Qualitative Social Research, 12(1), http://nbn-resolving.de/urn:nbn:de:0114-fqs1101C27
[Accessed: December 22, 2016].
Shabatay, Virginia (1991). The stranger's story: Who calls and who answers?.In Carol Witherell &
Nel Noddings (Eds.), Stories lives tell: Narrative and dialogue in education (pp.136-152). New York:
Teacher's College Press.
Sohn, Brian (2016). The student experience of other students. Doctoral dissertation, University of
Tennessee, Knoxville, USA, http://trace.tennessee.edu/utk_graddiss/3748/ [Accessed: June 20,
2016].
Sohn, Brian; Thomas, Sandra; Greenberg, Katherine & Pollio, Howard (under review). Hearing the
voices of students and teachers: A phenomenological approach to educational research.
Qualitative Research in Education.
Sohn, Brian; Plaas, Kristina; Franklin, Karen; Dellard, Tiffany; Murphy, Brenda; Greenberg,
Katherine; Greenberg, Neil; Pollio, Howard & Thomas, Sandra (2016). Freedom to connect: The
existential dimension of transformative learning in a graduate seminar. The Journal of
Transformative Education, 14(3), 178-199. doi: 10.1177/1541344616631425
Thomas, Sandra P. & Pollio, Howard R. (2002). Listening to patients: A phenomenological
approach to nursing research and practice. New York: Springer.
van Manen, Max (2014). Phenomenology of practice: Meaning-giving methods in
phenomenological research and writing. Walnut Creek, CA: Left Coast Press.
Woods, Megan; Paulus, Trena; Atkins, David & Macklin, Rob (2015). Advancing qualitative
research using qualitative data analysis software (QDAS)? Reviewing potential versus practice in
published studies using ATLAS.ti and NVivo, 1994–2013. Social Science Computer Review, 34(5),
597-617.
Author
Brian Kelleher SOHN is an adjunct professor in
the University of Tennessee, Knoxville's
Department of Educational Psychology and
Counseling. His research interests include
phenomenology, teaching and learning, and
classroom climate.
Contact:
Brian Sohn
535 Jane and David Bailey Education Complex
1122 Volunteer Boulevard
Knoxville, Tennessee 37996
USA
Tel: 1-865-974-8145
E-mail: bsohn@vols.utk.edu
FQS http://www.qualitative-research.net/
FQS 18(1), Art. 14, Brian Kelleher Sohn: Phenomenology and Qualitative Data Analysis Software (QDAS):
A Careful Reconciliation
Citation
Sohn, Brian Kelleher (2017). Phenomenology and Qualitative Data Analysis Software (QDAS): A
Careful Reconciliation [55 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative
Social Research, 18(1), Art. 14,
http://nbn-resolving.de/urn:nbn:de:0114-fqs1701142.
FQS http://www.qualitative-research.net/