Você está na página 1de 10

WD5891.

169-178 3/10/97 10:32 AM Page 169

Research in Nursing & Health, 1997, 20, 169–177

Focus on Qualitative Methods


Interpretive Description:
A Noncategorical Qualitative
Alternative for Developing
Nursing Knowledge
Sally Thorne,1* Sheryl Reimer Kirkham,1† Janet MacDonald-Emes2‡

1University
of British Columbia School of Nursing, 2211 Wesbrook Mall,
Vancouver, British Columbia V6T 2B5, Canada
2Emergency Department, Royal Victoria Hospital, Montreal, Quebec, Canada

Received 23 January 1996; accepted 28 August 1996

Abstract: Despite nursing’s enthusiastic endorsement of the applicability of qualitative research


approaches to answering relevant clinical questions, many nurse researchers have been hesi-
tant to depart from traditional qualitative research methods. While various derivations of phe-
nomenology, grounded theory, and ethnography have been popularized within qualitative nurs-
ing research, the methodological principles upon which these approaches are based reflect the
foundations and objectives of disciplines whose aims are sometimes quite distinct from nursing’s
domain of inquiry. Thus, as many nurse researchers have discovered, nursing’s unique knowl-
edge mandate may not always be well served by strict adherence to traditional methods as the
“gold standard” for qualitative nursing research. The authors present the point of view that a non-
categorical description, drawing on principles grounded in nursing’s epistemological mandate,
may be an appropriate methodological alternative for credible research toward the development
of nursing science. They propose a coherent set of strategies for conceptual orientation, sam-
pling, data construction, analysis, and reporting by which nurses can use an interpretive de-
scriptive approach to develop knowledge about human health and illness experience phenome-
na without sacrificing the theoretical or methodological integrity that the traditional qualitative
approaches provide. q 1997 John Wiley & Sons, Inc. Res Nurs Health 20: 169–177, 1997

Keywords: research method; qualitative research; interpretive description

The history and tradition within qualitative the unique demands of nurse researchers. We ex-
nursing research originates in the methodologies amine some of the reasons that nurse researchers
of several different disciplinary traditions. Just as have felt compelled to depart from traditional
quantitative scientific approaches proved insuffi- methods, despite a concurrent concern for retain-
cient for answering all of nursing’s theoretical and ing methodological integrity in their work. We
practical questions, the qualitative approaches de- take the position that it may be an appropriate time
rived from other disciplines have not always met in our history to consider noncategorical qualita-

This article is part of the ongoing series Focus on Qualitative Methods, edited by Margarete
Sandelowski, University of North Carolina, Chapel Hill.
Correspondence to Sally Thorne (thorne@nursing.ubc.ca).
*Associate professor.
†Doctoral student.
‡Staff nurse.

© 1997 John Wiley & Sons, Inc. CCC 0160-6891/97/020169-09 169


WD5891.169-178 3/10/97 10:32 AM Page 170

170 RESEARCH IN NURSING & HEALTH

tive research approaches that are derived from an the crudest form of inquiry, and even its rules were
understanding of nursing’s philosophical and the- dominated by the same fundamental assumptions.
oretical foundations as credible and legitimate Thus, for example, a description in this tradition
ways to access knowledge for nursing. In this con- was inherently better if it represented large rather
text, we propose interpretive description as one than small numbers of cases, if the sample ex-
such method. cluded rather than included outliers, and if rigid
strategies were used to ensure that data gathered
from all subjects were as identical as possible.
On the basis of these principles, the better data
LIMITATIONS IN TRADITIONAL
were those that could be quantified and, therefore,
(QUANTITATIVE) SCIENCE reduced to mathematical probability logic (Lein-
inger, 1985). Having followed all of these rules,
To a considerable extent, the popularity of quali- high quality description would typically be decon-
tative research approaches within nursing science textualized to the point that it was almost devoid
can be attributed to nursing’s increasingly confi- of human subjectivity. If qualitative (“soft”) data
dent critique of the limits of traditional science for were included in a final report, they would be re-
developing the kinds of knowledge that are re- ported almost apologetically, with care to avoid
quired for nursing practice. Nursing’s practice any possible accusation that the researcher had
questions typically challenge us in the direction of been influenced (“biased”) by them in the analyt-
two equally compelling dimensions: knowing that ic process. Thus, attempts to answer nursing’s
which is shared by persons in similar situations, questions about health and illness experience
and knowing that which is particular to the lived within the quantitative descriptive tradition were
experience of an individual person (Colaizzi, somewhat limited in their scope and depth, and did
1975; Coward, 1990). Because traditional science not always satisfy the requirements of a holistic,
orients itself entirely toward the shared compo- interpretive, relational practice discipline.
nents of experience, it permits searching for pop-
ulation patterns, correlations, and tendencies
among aggregates, redirecting our focus away LIMITATIONS IN THE QUALITATIVE
from a sense of individuals in context (Dzurec, TRADITION
1989). In the search for knowledge that can be ap-
plied in the practice context, nurses have demand- Despite a passion for engaging in describing ele-
ed forms of inquiry that reveal processes for ap- ments of the human condition in health and illness,
plying aggregated knowledge to individual cases. pioneering qualitative researchers were under-
Nursing theory has been one mechanism facilitat- standably reluctant to align themselves with the
ing such inquiry (Mitchell & Cody, 1992). descriptive tradition of quantitative scientific
At the same time, a forceful critique of normal methods. They astutely recognized that the most
science, in general and in the human health field, effective way to distance themselves from this tra-
has created a climate in which researchers are in- dition was to locate their science within the legit-
creasingly aware that quantitative approaches sys- imized philosophical and methodological projects
tematically discount certain species of knowledge, of other disciplines (Morse, 1994b). As has been
reveal assumptions about essential truths that may explained elsewhere (Thorne, 1991), these quali-
not be shared by all in the discipline, and rely on tative nurse researchers generally sought episte-
rules of science that are dependent upon those pat- mological credibility in three primary directions:
terns and assumptions (Newman, 1992). As a con- the phenomenological project within philosophy,
sequence, a climate more tolerant or supportive of the grounded social theory project within sociolo-
qualitative research approaches has emerged in gy, and the ethnographic project within cultural
the health sciences where it did not exist in previ- anthropology. Formal study of any of these three
ous decades (Kidd & Morrison, 1988). disciplinary traditions reveals that there are com-
Because of the dominance of normal science plex relationships between the methodological
approaches within the culture of nursing acade- standards and the larger objectives of the disci-
mia, the early qualitative nursing researchers were pline (Atkinson, 1995). For example, ethnogra-
compelled to engage in elaborate defenses of the phy’s rules derive not only from the desire to doc-
theoretical and methodological foundations of ument human variation, but also from the passion
their research methods (Leininger, 1985). It for discovering human universals. Phenomenolo-
should be remembered that, within the traditional gy’s methods assume the general philosophical
empirical science domain, description served as stance that there is essential structure to human ex-
WD5891.169-178 3/10/97 10:32 AM Page 171

INTERPRETIVE DESCRIPTION / THORNE ET AL. 171

perience. Grounded theory methodology is depen- Parse’s methodology (1990)], they tend to be quite
dent upon the assumption that human social restricted in focus and aligned with the advance-
processes beyond individual consciousness con- ment of one or another theoretical position. How-
strain and explain human behavior. In order to ever, as Morse (1989a) points out, there are a num-
place their research within the context of estab- ber of nurses who are doing legitimate qualitative
lished scientific inquiry, most early qualitative research for which there is as yet no name. Often,
nursing researchers of any caliber aligned with the such research involves description of and inter-
coherent logic of one or another of these ap- pretation about a shared health or illness phenom-
proaches, as is evidenced by a generation of nurse enon from the perspective of those who live it.
phenomenologists, ethnographers, and grounded Such descriptions can be considered a quintessen-
theorists (Mitchell & Cody, 1993). tially nursing form of science in that they reflect a
Within the academic establishment, method- respect for knowledge about aggregates in a man-
ological variation from these traditions by nurse ner that does not render the individual case invis-
researchers was not encouraged. When it oc- ible.
curred, it was often demeaned as “mixed meth- In this more eclectic and less rigid disciplinary
ods” (Leininger, 1992; Morse, 1989a), “method academic climate, nurse scholars have also dis-
slurring” (Baker, Wuest, & Stern, 1992), and in covered that postmodern thinking provides a chal-
general considered “sloppy” research (Morse, lenge to traditional assumptions about “truth”
1989a). Indeed, a researcher claiming to “do phe- within all of the sciences, and provides a broad
nomenology” to reveal lived experience but using foundation for inquiry that respects the dialectic
ethnographic interview methods and applying between the general and particular, between com-
grounded theory analytic approaches such as con- monality and individuality, between truth and per-
stant comparative analysis did seem to have ception, between theory and practice (Moccia,
missed the point and thereupon joined the “min- 1988; Watson, 1995). Unlike many other sciences,
ions of mediocrity” (Stern, 1994). nursing has the distinct advantage of being an ap-
However, a generation of qualitative nurse re- plied or practical science, and, therefore, nurse
searchers has learned its craft outside of the con- theorists have never assumed that their scholar-
strained academic climate forced upon its prede- ship was simply theoretical. Concurrent with a sci-
cessors. Because of the interdisciplinarity inherent ence of the general, practice scholars have gener-
in nursing science and the availability of doctoral ated a significant body of work on the particular
study in nursing, nurses are less commonly (how nurses come to “know” their patients, how
pressed into the disciplinary alliances of those intuition and pattern recognition develop in expert
who took their doctorates in philosophy, sociolo- practice, how unrestricted awareness of possibili-
gy, or anthropology. Therefore, they may feel ties enhances the processes of clinical reasoning,
more freedom to examine methodological ques- how nurses practice the art of nursing). Thus, we
tions in the context of the overall objectives of have a practice and scholarship climate within
nursing science instead of following the method- nursing that is clamoring for general knowledge of
ological dictates of other disciplines. The nature of the sort that enhances particularization in practice
nursing knowledge is such that nurse researchers (Dzurec, 1989).
often find departures from the traditional qualita- The qualitative species of knowledge that nurs-
tive approaches to be necessary and appropriate ing practice theory demands can be quite different
(Dreher, 1994). Seeking methods that are specifi- in its nature from the kind of knowledge that tra-
cally suited to answering the kinds of research ditional quantitative descriptive research was de-
questions that will advance nursing theory and sci- signed to access (Carter, 1985). To illustrate,
ence, they have made considerable advances in the where a quantitative descriptive study would have
development of some of these traditional qualita- evaluated levels or correlates of anxiety in relation
tive methodological approaches. Some examples to a particular diagnostic test, the qualitative re-
include the work of Sandelowski (1991) in narra- searcher might explore ways in which that anxiety
tive analysis, Swanson and Chapman (1994) in was manifested, subjective perceptions about its
qualitative evaluation, and Carey (1994) in focus origins, or patterns with which it might be intensi-
group research. In most instances, these efforts fied or alleviated throughout the procedure. The
have not been articulated as nursing science quantitative descriptive approach assumed that a
methodologies but rather as adaptations of the fundamental law of nature explained elements of
methods pioneered within other disciplines. the phenomenon of anxiety in diagnostic testing,
Where methods are depicted as unique to nurs- and that the descriptive findings would bring sci-
ing [such as Leininger’s ethnonursing (1991) and ence one step closer to identifying and classifying
WD5891.169-178 3/10/97 10:32 AM Page 172

172 RESEARCH IN NURSING & HEALTH

that law (Packard & Polifroni, 1992). In direct quirements for trustworthy research reports (Lin-
contrast, a qualitative descriptive approach would coln & Guba, 1985; Sandelowski, 1986). Without
assume that, while patterns within human behavior them, we can justly be accused of sloppy science.
might be explicable using one or another theoret- It is apparent that simple recipes for qualitative re-
ical proposition, recognition that they might occur search will not replace sound methodological rea-
was more important than explanation in the clini- soning, and qualitative researchers will continue
cal application domain. Thus, qualitative research- to require solid theoretical foundations upon
ers maintained their distance from traditional de- which to build new or adapted research approach-
scription while, at the same time, loosening the es. Instead of claiming a hollow allegiance to the
bonds that tied them to the rigid methodologies of accepted methodological positions when their
the other disciplines. However, unlike researchers work reflects a uniquely nursing adaptation, we
in the field of education, who have unselfcon- strongly encourage qualitative nurse researchers
sciously developed and named their own qualita- to make explicit their departures from tradition, to
tive research traditions (such as the naturalistic in- name them as distinct methodological approaches
quiry of Lincoln and Guba, 1985) for their own and, thereby, to begin the process of legitimizing
unique disciplinary projects (such as evaluating them within our scholarly discourse.
programs), qualitative research approaches tai- As nurses push past the limits of traditional
lored to nursing’s distinctive aims have not yet methods, embrace new possibilities, and create
been articulated. It is our view that nursing’s new blends of strategy, their thoughtful attention
uniqueness has gradually shifted the priorities to the theoretical traditions from which the com-
within our research enterprise to the point that we mon qualitative methods were derived and the
can begin to build methods that are grounded in philosophical claims upon which nursing is based
our own epistemological foundations, adhere to can help them build a distinctly nursing research
the systematic reasoning of our own discipline, logic for studying human health and illness expe-
and yield legitimate knowledge for our practice. rience questions. Among the foundations of nurs-
We put forward interpretive description as one of ing knowledge underlying such methods will be a
many possible “generic” nursing approaches. recognition that human health and illness experi-
ences are comprised of complex interactions be-
tween psychosocial and biological phenomena,
that common patterns within such experiences
METHODOLOGICAL APPLICATIONS
represent the core of our disciplinary practice
FOR NURSING RESEARCH knowledge, and that the practical application of
principles derived from such common patterns
Because nurse researchers increasingly are drawn will always be individualizable in the context of a
by virtue of their understanding of nursing knowl- particular case. In contrast to its logical empiricist
edge to methodological applications that are not epistemological origins, today’s nursing science
directly borrowed from one or another of the pre- seeks as its “truths” a set of ideas that have appli-
dominant qualitative research traditions, it is im- cation potential, but remain amenable to reconsid-
portant that a body of scholarship be developed eration in the light of varying contexts, new con-
and applied to the question of standards and cred- cepts, new ways of understanding, and new
ibility measures for such research. Those schooled meanings. The qualitative nursing research ap-
in traditional qualitative approaches will recog- proach suggested here is grounded in an interpre-
nize that respect for the philosophical and disci- tive orientation that acknowledges the constructed
plinary traditions has been an integral aspect of and contextual nature of much of the health–ill-
sound methodological reasoning within our nurs- ness experience, yet also allows for shared reali-
ing qualitative research tradition. In fact, we might ties. As such, it differs from eclectic approaches
collectively be accused of an obsession about that “slur” methods without regard for the coher-
methodological integrity—what Janesick (1994) ence of their epistemological foundations.
refers to as methodolatry, or what Atkinson (1995)
has called fetishizing method.
Although there have been some attempts with-
in our literature to simplify and codify qualitative INTERPRETIVE DESCRIPTION OF
methods (e.g., Strauss & Corbin, 1990), we have HEALTH AND ILLNESS EXPERIENCES
generally adopted the understanding that coherent
logic within the analytic frame and a traceable au- In this spirit, we put forth an argument for inter-
dit trail for the inductive reasoning process are re- pretive description as one approach that can be ap-
WD5891.169-178 3/10/97 10:32 AM Page 173

INTERPRETIVE DESCRIPTION / THORNE ET AL. 173

plied to qualitative inquiry into human health and for what is found in the inquiry, it typically will be
illness experiences for the purpose of developing challenged as the inductive analysis proceeds.
nursing knowledge. We believe that qualitative However, because it is explicit in the description
nursing research following certain general princi- of the work, it also will provide a solid basis upon
ples for analytic frameworks, sample selection, which the design logic and the inductive reason-
data sources, data analysis, and rigor represents a ing in interpreting meanings within the data can be
credible means by which to develop clinical judged (May, 1989). As time passes and new
knowledge with significant nursing science appli- meanings emerge in our theoretical literature, an
cations. Therefore, we present considerations of explicit analytic forestructure in the records of our
each of these components of qualitative inquiry in research will permit nurse scholars to make sense
a manner that we believe reflects a solid ground- of the findings and develop increasingly complex
ing in the nature of practice knowledge and nurs- interpretations of how they contribute to our sci-
ing science. Following the logic of these general ence.
principles, we believe that nurses can create sound
interpretive description that contributes directly to Sample Selection
our understanding of how people experience their
health and illness and what nursing can do to make The general principle of theoretical sampling
a difference. In our view, interpretive descriptions makes a useful contribution to the design of an in-
of increasing complexity and interrelatedness rep- terpretive description for the purposes of nursing
resent the foundation for nursing’s theoretical knowledge development. In its most simplistic
structure and its substantive body of knowledge. form, theoretical sampling encourages us to sam-
ple from the most predictable variations within the
Analytic Frameworks theme we are studying (Morse, 1995). However,
much of the available qualitative research is quite
In contrast to traditional phenomenological in- limited in its appreciation for the variables on
quiry, nursing’s interpretive description ought to which such sampling is conducted. While an ini-
be located within the existing knowledge so that tial analytic framework can provide some useful
findings can be constructed on the basis of direction for preliminary decisions, we would ar-
thoughtful linkages to the work of others in the gue that the principle is most appropriately applied
field (Mitchell & Cody, 1993). In the past, some to the notion of obtaining maximal variation on the
qualitative nurse researchers have defended their themes that emerge from the inductive analysis it-
absence of a theoretical foundation with the claim self (Glaser, 1978; Sandelowski, 1995). Usually,
that “nothing is known” about certain illness ex- the positions or experiences that each participant
periences. While formal research into a phenome- or informant might represent cannot be known un-
non might not have been published, such claims til data collection is well underway. For example,
typically ignore a body of clinical knowledge that while our initial framework might have told us that
may have equal value. Therefore, for the purpos- married and single subjects could have different
es of an interpretive description, we suggest that experiences in relation to emotional support, our
what is known, whether by virtue of formal re- developing analysis might suggest that various
search or of clinical interpretation, should be con- gender role assumptions within marriage repre-
sidered foundational forestructure to a new in- sent an equally critical variable determining
quiry (Schultz & Meleis, 1988). Like Morse whether a woman is comfortable in seeking sup-
(1994a), we argue that “going in blind” can be port outside the marriage. We might then actively
counterproductive to nursing’s scientific knowl- sample among women with varying views on their
edge development. However, in contrast to tradi- role within the marriage in order to develop a more
tional descriptive research, in which a formal con- complex interpretation of patterns relevant to
ceptual framework would be required, an analytic emotional support.
framework constructed on the basis of critical Further, caution must be applied to the use of
analysis of the existing knowledge represents an single representatives of a specific position in our
appropriate platform on which to build a qualita- attempts to achieve maximal variation on any
tive design. Such a framework orients the inquiry, theme. Because no research subject ever repre-
provides a rationale for its anticipated boundaries, sents the essence of a single variable and none oth-
and makes explicit the theoretical assumptions, bi- er, serious errors can be made if we misinterpret
ases, and preconceptions that will drive the design the contributions of individual participants. Fol-
decisions (May, 1989). Because it represents a be- lowing on the example given above, a married
ginning point rather than an organizing structure woman who overtly ascribes to traditional gender
WD5891.169-178 3/10/97 10:32 AM Page 174

174 RESEARCH IN NURSING & HEALTH

relations within marriage may well feel absolute Intensive interviewing and participant observa-
control over her own choices in life, while a tion are notoriously time consuming and expen-
woman claiming hers to be a “liberated” marital sive. Despite this, many qualitative nursing stud-
relationship may crave the passive dependence ies rely exclusively on these techniques
portrayed in romance novels. In order to do justice (Hutchinson & Wilson, 1994; May, 1989). For the
to a description that has explanatory power for un- purposes of interpretive description, we believe
derstanding what might be shared or common that appropriate collateral data sources often are
within a phenomenon such as emotional support, available for qualitative nurse researchers inter-
the researcher would need a sufficiently strong ested in expanding the scope of their inquiry,
database upon which to make confident claims broadening the reach of their theoretical sample,
about commonalities and differences across all of without incurring excessive costs. In relation to
the variables that are central to the analysis many health and illness experiences, a body of lay
(Morse, 1995). If such a foundation is not possible print or other media information as well as nurs-
(as in a small preliminary study), understanding ing case reports and clinical papers is often readi-
the principle of theoretical sampling can guide us ly available, and can provide a strong backbone of
in articulation of probable limitations to how we support for qualitative nursing inquiries. At the
might interpret the meaning of the findings. very least, such materials can provide a testing
ground for the developing insights that emerge in
Data Sources the data analysis. If the conceptualizations would
not match the anecdotal reports beyond the influ-
In keeping with many qualitative researchers, we ence of the researcher and the data-gathering pro-
contend that people who have lived with certain cedures, then the researcher ought to be able to ex-
experiences are often the best source of expert plain why not. Thus, we believe that the judicious
knowledge about those experiences (Morse, application of a range of data sources can add con-
1989b). However, we would also argue that not all siderable strength to the usual data sources of in-
people who have such experiences will make good terviews and observations for the purposes of gen-
research participants (Morse, 1989a). In our expe- erating practice knowledge for nursing.
rience, some subjects are articulate, thoughtful,
and eager to share their abstractions and analyses Data Analysis
of a situation, while others tend to be more con-
crete and more comfortable with events than inter- In the qualitative tradition, inductive rather than
pretations. Complicating the situation further, it is deductive analysis is required and, therefore, we
important to recognize that many people weave recommend that techniques that inhibit the former
their recollections of subjective experience into are generally to be avoided. Examples of deduc-
the preexisting tapestry that is their life narrative. tively derived analytic techniques might include
Such narratives may take on an infinite number of predetermined analytic strategies, such as content
themes, but some common ones that may be fa- analysis, and overly small units of analysis, such
miliar to the reader include “life has never been as words or phrases. Premature coding, arising
fair to me,” “behind every cloud is a silver lining,” from an eagerness not to let data gathering get out
or “suffering always makes us strong.” It can be- of hand, can privilege superficial understandings
come almost impossible for a researcher to untan- at the expense of deeper and more meaningful an-
gle the shared component of a subjective experi- alytic interpretations. Further, it is our view that
ence from the narratives that people place them in, complex coding systems, such as those that en-
and so qualitative research tends to require courage multiple codings for all pieces of raw
thoughtful analysis of the relationship between the data, often overwhelm the researcher with detail to
data sources and the findings that derive from the point that inductive interpretation becomes al-
them. Thus, we believe that an interpretive de- most inconceivable. Lowenberg (1993) attributes
scription that is meant to generate nursing practice these problems to an emphasis on the technical
knowledge will require purposeful selection of re- rather than the theoretical or epistemological as-
search participants whose accounts reveal ele- pects of the method. From our perspective, strug-
ments that are to some degree shared by others. gling to apprehend the overall picture with ques-
Not all of the data derived from those who have ex- tions such as “what is happening here?” and “what
perienced a phenomenon will reflect this common am I learning about this?” will typically stimulate
nature, and an effective interpretive description more coherent analytic frameworks for interpre-
will be one that distinguishes eccentricities from tive description than will sorting, filing, and com-
commonalities within its process and outcome. bining vast quantities of small data units, regard-
WD5891.169-178 3/10/97 10:32 AM Page 175

INTERPRETIVE DESCRIPTION / THORNE ET AL. 175

less of the mechanical techniques or computer visible as factors in shaping our research interpre-
software employed. Effective qualitative research tations. However, their influence upon the ques-
requires endurance and patience, and it is the rare tions we ask, the way we ask them, and the meth-
researcher who will be able to discard an exten- ods by which we seek to answer them are
sive data sorting scheme because a more appro- undeniable.
priate abstraction to represent the whole becomes Many researchers find a reflective journal a
apparent. valuable asset to guiding as well as documenting
Thus, we would recommend analytic tech- the reactive processes of interpreting or counter-
niques such as the adaptations explicated by Gior- ing bias within the research process (Lamb & Hut-
gi (1985), Knafl and Webster (1988), or Lincoln tlinger, 1989; Lincoln & Guba, 1985; Paterson,
and Guba (1985) that encourage repeated immer- 1994). Field notes that record the context of all
sion in the data prior to beginning coding, classi- data-gathering episodes and link those contexts to
fying, or creating linkages. These analytic proce- the phenomena under study are also extremely
dures capitalize on such processes as synthesizing, useful (Dreher, 1994). In our view, an interpretive
theorizing, and recontextualizing rather than sim- description will require some such means by
ply sorting and coding (Morse, 1994b). Further, which to retrace the development of abstractions
strategic periods of immersion in the field inter- and to ensure that the analytic directions are de-
spersed with periods of immersion in the data fensible. While the limits of most scholarly publi-
seem ideally suited to research endeavors that re- cations preclude detailed descriptions of methods
quire refining the inquiry, testing the developing (Thorne, 1994), sufficient information must be
conceptualizations, and challenging the abstrac- available in research reports for readers to follow
tions that emerge with strategic theoretical sam- the analytic reasoning process and to judge the de-
pling (Lofland, 1976; Strauss, 1987). Because gree to which the analysis is grounded within the
they address the dialectic between individual cas- data. For example, the logic of theoretical sam-
es and common patterns, nursing studies can also pling and the variables about which the researcher
capitalize on this strategy. Interpretive description makes confident claims must be explicit in the re-
in nursing requires that nurse researchers come to port to make an interpretive description of the phe-
know individual cases intimately, abstract relevant nomenon convincing.
common themes from within these individual cas- While attention to such process issues creates
es, and produce a species of knowledge that will some confidence in the findings, the truth value
itself be applied back to individual cases. In order (or theoretical validity) of qualitative research re-
to do this effectively, they must engage in both the sults also requires that steps be taken to ensure that
ethereal abstractions of theorizing and the earth- researcher bias or overenthusiasm has not system-
bound concrete realities of the practice context in atically skewed the findings of the study (Brink,
order to produce sound and usable knowledge. 1989). Typically, repeated interviewing, in which
developing conceptualizations can be subjected to
Rigor challenges or refinements is built into the design
of an interpretive description. In our view, taking
Because the design in qualitative research will the raw data (such as transcripts) back to partici-
necessarily be somewhat emergent (Brink & pants for a credibility check is generally insuffi-
Wood, 1989; Sandelowski, Davis, & Harris, cient for these purposes and may, as Sandelowski
1989), attention to rigor in the process and the re- (1993) points out, create contradictions within the
porting of that process is critical to an interpretive process of developing knowledge. Instead, it is our
description. Attempts to eliminate all biases are view that beginning conceptualizations, repre-
naive; therefore, the researcher must explicitly ac- senting the entire sample rather than the individ-
count for the influence of bias upon the research ual research subject, are more usefully brought to
findings as much as possible (Dreher, 1994). In ad- individual research participants for their critical
dition to the individual or substantive biases that a consideration. Often, more important insights
researcher may bring to a study, we believe that about a conceptualization can be formulated from
there may be inherent biases in research that takes people’s perceptions of why it does not quite fit
a nursing perspective. Such biases might include than why it does! Such a strategy creates optimal
convictions about the value of a common social conditions for challenging the emergent theoriz-
good, the belief that suffering can be ameliorated, ing and refining the theoretical linkages. Because
or the view that all people are deserving of the re- of this, it permits the nurse researcher to come
sources for health. Because they are so integral to away with confidence that the conceptualizations
nursing philosophy, such perspectives can be in- are, indeed, grounded in data and representative of
WD5891.169-178 3/10/97 10:32 AM Page 176

176 RESEARCH IN NURSING & HEALTH

shared realities rather than an artifact of design or Baker, C., Wuest, J., & Stern, P.N. (1992). Method slur-
instrument (researcher) error. From our perspec- ring: The grounded theory/phenomenology example.
tive, a caring discipline such as nursing has a spe- Journal of Advanced Nursing, 17, 1355–1360.
cial obligation to ensure that the rigor of its re- Brink, P.J. (1989). Issues in reliability and validity. In
search findings are above reproach. While our J.M. Morse (Ed.), Qualitative nursing research: A
contemporary dialogue (pp. 151–168). Rockville,
inherent appreciation for subjectivity can be a MD: Aspen.
tremendous asset in our practice as well as our sci- Brink, P., & Wood, M. (1989). Advanced design in nurs-
ence, we cannot fulfill our social mandate on the ing research. London: Sage.
basis of a scholarship that ignores objectivity and Carey, M.A. (1994). The group effect in focus groups:
competing truth claims. Planning, implementing and interpreting focus group
research. In J.M. Morse (Ed.), Critical issues in qual-
itative research methods (pp. 225–241). Thousand
CONCLUSION Oaks, CA: Sage.
Carter, M.A. (1985). The philosophical dimensions of
qualitative nursing science research. In M.M.
We have taken the view that there is considerable Leininger (Ed.), Qualitative research methods in
room for nursing to advance qualitative research nursing (pp. 27–32). Orlando, FL: Grune & Stratton.
methodology beyond the approaches designed for Colaizzi, J. (1975). The proper object of nursing sci-
the purposes of other academic disciplines and to ence. International Journal of Nursing Studies, 12,
develop its own distinct traditions and conven- 197–200.
tions. Although our literature reveals that nurse re- Coward, D.D. (1990). Critical multiplism: A research
searchers have creatively applied established strategy for nursing science. Image: Journal of Nurs-
methods from other disciplines, it seems that they ing Scholarship, 22, 163–167.
Dreher, M. (1994). Qualitative research methods from
have been somewhat reluctant to claim and name the reviewer’s perspective. In J. Morse (Ed.), Critical
qualitative research approaches that reflect a dis- issues in qualitative research methods (pp. 281–297).
tinctly nursing orientation. In keeping with the Thousand Oaks: Sage.
work of those in other disciplines to shape new Dzurec, L.C. (1989). The necessity for and evolution of
conventions within the interpretivist and natural- multiple paradigms for nursing research: A poststruc-
istic traditions (Guba, 1990; Lowenberg, 1993; turalist perspective. Advances in Nursing Science,
Schwandt, 1994), it seems time to recognize our 11(4), 69–77.
collective strengths in this sort of enterprise. By Giorgi, A. (1985). Phenomenology and psychological
virtue of the nature of the phenomena in which research. Pittsburgh: Duquesne University Press.
nursing is interested as well as nursing’s practice Glaser, B.G. (1978). Theoretical sensitivity. Mill Valley,
CA: Sociology Press.
mandate, nurse researchers are ideally placed to Guba, E. (1990). The paradigm dialog. Newbury Park,
thoughtfully modify established methods and de- CA: Sage.
velop inquiry approaches that better align with our Hutchinson, S., & Wilson, H. (1994). Research and ther-
unique philosophical foundations, interpretive apeutic interviews: A poststructuralist perspective. In
themes, and disciplinary objectives. Interpretive J.M. Morse (Ed.), Critical issues in qualitative re-
description may be one such means by which nurs- search methods (pp. 300–315). Thousand Oaks, CA:
es can derive clinical knowledge applicable to de- Sage.
veloping their practice science. We have articulat- Janesick, V.J. (1994). The dance of qualitative research
ed general principles of an interpretive descriptive design: Metaphor, methodolatry, and meaning. In
approach that we believe reflects nursing’s unique N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of
qualitative research (pp. 209–219). Thousand Oaks,
mandate and epistemological foundations. It is our CA: Sage.
hope that, in their quest to develop practice knowl- Kidd, P., & Morrison, E.F. (1988). The progression of
edge, qualitative nurse researchers will continue knowledge in nursing: A search for meaning. Image:
to develop rigorous and credible methodologic Journal of Nursing Scholarship, 20, 222–224.
adaptations for which the gold standard is not sim- Knafl, K.A., & Webster, D.C. (1988). Managing and an-
ply adherence to the rules of a traditional method alyzing qualitative data: A description of tasks, tech-
but a coherent logic derived from the philosophy niques, and materials. Western Journal of Nursing
and science of nursing. Research, 10, 195–218.
Lamb, G., & Huttlinger, K. (1989). Reflexivity in nurs-
ing research. Western Journal of Nursing Research,
11, 765–772.
REFERENCES Leininger, M. (1985). Nature, rationale, and importance
of qualitative research methods in nursing. In M.M.
Atkinson, P. (1995). Some perils of paradigms. Quali- Leininger (Ed.), Qualitative research methods in
tative Health Research, 5, 117–124. nursing (pp. 1–25). Orlando, FL: Grune & Stratton.
WD5891.169-178 3/10/97 10:32 AM Page 177

INTERPRETIVE DESCRIPTION / THORNE ET AL. 177

Leininger, M.M. (1991). Ethnonursing: A research Parse, R.R. (1990). Parse’s research methodology with
method with enablers to study the theory of culture an illustration of the lived experience of hope. Nurs-
care. In M.M. Leininger (Ed.), Culture care diversity ing Science Quarterly, 3, 9–17.
and universality: A theory of nursing (pp. 73–117). Paterson, B.L. (1994). A framework to identify reactiv-
New York: National League for Nursing Press. ity in qualitative research. Western Journal of Nurs-
Leininger, M. (1992). Current issues, problems and ing Research, 16, 310–316.
trends to advance qualitative paradigmatic research Sandelowski, M. (1986). The problem of rigor in qual-
methods for the future. Qualitative Health Research, itative research. Advances in Nursing Science, 8(3),
2, 392–415. 27–37.
Lincoln, Y.S., & Guba, E.G. (1985). Naturalistic in- Sandelowski, M. (1991). Telling stories: Narrative ap-
quiry. Beverly Hills: Sage. proaches in qualitative research. Image: Journal of
Lofland, J. (1976). Developing disciplined abstractions. Nursing Scholarship, 23, 161–166.
In J. Lofland, Doing social life: The qualitative study Sandelowski, M. (1993). Rigor and rigor mortis: The
of human interaction in natural settings (pp. 62–79). problem of rigor in qualitative research revisited. Ad-
New York: Wiley. vances in Nursing Science, 16(2), 1–8.
Lowenberg, J.S. (1993). Interpretive research method- Sandelowski, M. (1995). Sample size in qualitative re-
ology: Broadening the dialogue. Advances in Nursing search. Research in Nursing & Health, 18, 179–183.
Science, 16(2), 57–69. Sandelowski, M., Davis, D.H., & Harris, B.G. (1989).
May, K. (1989). Interview techniques in qualitative re- Artful design: Writing the proposal for research in the
search: Concerns and challenges. In J. Morse (Ed.), naturalist paradigm. Research in Nursing & Health,
Qualitative nursing research: A contemporary dia- 12, 77–84.
logue (pp. 188–201). Newbury Park, CA: Sage. Schultz, P.R., & Meleis, A.I. (1988). Nursing episte-
Mitchell, G.J., & Cody, W.K. (1992). Nursing knowl- mology: Traditions, insights, questions. Image: Jour-
edge and human science: Ontological and epistemo- nal of Nursing Scholarship, 20, 217–221.
logical considerations. Nursing Science Quarterly, 5, Schwandt, T.A. (1994). Constructivist, interpretivist ap-
54–61. proaches to human inquiry. In N.K. Denzin & Y.S.
Mitchell, G.J., & Cody, W.K. (1993). The role of theo- Lincoln (Eds.), Handbook of qualitative research
ry in qualitative research. Nursing Science Quarter- (pp. 118–137). Thousand Oaks, CA: Sage.
ly, 6, 170–178. Stern, P. (1994). Eroding grounded theory. In J. Morse
Moccia, P. (1988). A critique of compromise: Beyond (Ed.), Critical issues in qualitative research methods
the methods debate. Advances in Nursing Science, (pp. 212–223). Thousand Oaks, CA: Sage.
10(4), 1–9. Strauss, A.L. (1987). Qualitative analysis for social sci-
Morse, J.M. (1989a). Qualitative nursing research: A entists. New York: Cambridge University Press.
free-for-all? In J.M. Morse (Ed.), Qualitative nursing Strauss, A., & Corbin, J. (1990). Basics of qualitative re-
research: A contemporary dialogue (pp. 3–10). search: Grounded theory procedures and techniques.
Rockville, MD: Aspen. Newbury Park, CA: Sage.
Morse, J.M. (1989b). Strategies for sampling. In J.M. Swanson, J.M., & Chapman, L. (1994). Inside the black
Morse (Ed.), Qualitative nursing research: A con- box: Theoretical and methodological issues in con-
temporary dialogue (pp. 117–131). Rockville, MD: ducting evaluation research using a qualitative ap-
Aspen. proach. In J.M. Morse (Ed.), Critical issues in quali-
Morse, J.M. (1994a). Going in “blind.” Qualitative tative research methods (pp. 66–93). Thousand Oaks,
Health Research, 4, 3–5. CA: Sage.
Morse, J. (1994b). Qualitative research: Fact or fanta- Thorne, S.E. (1991). Methodological orthodoxy in qual-
sy? In J. Morse (Ed.), Critical issues in qualitative re- itative nursing research: Analysis of the issues. Qual-
search methods (pp. 1–8). Thousand Oaks, CA: Sage. itative Health Research, 1, 178–199.
Morse, J.M. (1995). The significance of saturation. Thorne, S. (1994). Secondary analysis in qualitative re-
Qualitative Health Research, 5, 147–149. search: Issues and implications. In J.M. Morse (Ed.),
Newman, M.A. (1992). Prevailing paradigms in nurs- Critical issues in qualitative research methods
ing. Nursing Outlook, 40, 10–13, 32. (pp. 263–279). Thousand Oaks, CA: Sage.
Packard, S.A. & Polifroni, E.C. (1992). The nature of Watson, J. (1995). Postmodernism and knowledge de-
scientific truth. Nursing Science Quarterly, 5, velopment in nursing. Nursing Science Quarterly, 8,
158–163. 60–64.
WD5891.169-178 3/10/97 10:32 AM Page 178

Você também pode gostar