Vous êtes sur la page 1sur 8

Research in Nursing and Health, 1984, 7, 17-24

Interpreting and Reporting


Qualitative Research
Kathleen Astin Knafl and Marion J. Howard

The aims of this paper were to clarify the nature a n d purpose of qualitative research
and to suggest specific guidelines for reporting the results of qualitative studies. Con-
fusion about the purpose of qualitative research and the absence of a standardized
format for reporting such research make it difficult to assess the validity of qualitative
studies. Four general purposes of qualitative research were discussed: instrument de-
velopment, illustration of results, description, a n d conceptualization. A framework for
reporting qualitative research that provides a guide for assuring consistency across
original study purpose, study design, a n d final report was presented and can be used
for either evaluating or writing a report of qualitative research.

The reporting of study results confronts the of qualitative research. Building on these as-
qualitative researcher with a difficult problem. sumptions, the aims of this article are to clarify
Unlike the person who has undertaken an exper- the nature and purposes of qualitative research
imental or survey study, the qualitative re- and to suggest specific guidelines for reporting
searcher has no well-codified, generally ac- the results of qualitative studies.
cepted, protocol available as to how the
methodology and findings of such a study best
can be communicated. Focusing on techniques
NATURE AND PURPOSE OF QUALITATIVE
and problems of data collection, processing, and
RESEARCH
analysis, qualitative methods texts and articles
give the researcher comparatively few guidelines In this section the desired products of quali-
for communicating data in a way that convinces tative nursing research and the means for attain-
the reader of their validity and reliability. As a ing these are discussed. Qualitative research typ-
result, reports of qualitative studies follow vary- ically has been defined in terms of research
ing formats (Glaser & Strauss, 1966; Klenow, methods. Comparing quantitative and qualitative
1981), many of which may appear highly unor- research, Cook and Reichardt (1979) stated:
thodox and unacceptable to the reader who is
accustomed to reading the results of surveys and By quantitative methods, researchers have come
experiments. Using the latter as a standard for to mean the techniques of randomized exper-
evaluation, readers often find reports of qualita- iments, quasi-experiments, paper and pencil
tive research interesting but unconvincing. “objective”. tests, multivariate statistical anal-
The present article is predicated on the follow- yses, sample surveys, and the like. In contrast,
ing two assumptions: (a) the absence of a standard qualitative methods include ethnography, case
format for reporting qualitative research makes it study, indepth interviews, and participant ob-
difficult for even the methodologically sophisti- servation. (p. 7)
cated reader to assess the validity of a qualitative
study, and (b) difficulties in understanding and Qualitative research is equated with those
evaluating qualitative research stem from con- methods or data gathering techniques which gen-
fusion over the underlying nature and purposes erate narrative as opposed to numerical data.

Dr. Kathleen Knafl i s a professor in psychiatric nursing in the College of Nursing at the University
of Illinois-Chicago. Ms. Marion Howard i s a research associate at the Erie Family Health Center,
Inc., Chicago, Illinois.
This article was received June 7, 1982, wos revised, ond an December 10, 1982, was accepted
for publication.
Requests for reprints moy be addressed to Dr. Kathleen Astin Knafl, Professor, Psychiatric Nursing,
College of Nursing, University of Illinois-Chicago, 845 South Damen Avenue, Chicago, I1 60612.

0 1984 Wiley 01 60-6891/84/010017-08 $04.00 17


18 RESEARCH IN NURSING AND HEALTH

Qualitative data take the form of verbatim inter- related to purpose. For example, in a study of
view and/or field note transcripts. reality shock among new baccalaureate graduate
What the preceding distinctions fail to take into nurses in medical center hospitals, Kramer (1969)
account are the purposes served by qualitative reported the qualitative component of her study
methods and data gathering techniques. Ethnog- in grounded theory terms when she actually used
raphies, case studies, intensive interviews, and the qualitative data to illustrate predominantly
participant observation can serve a variety of re- quantitative findings. Other authors inadequately
search purposes. For example, nurse authors de- described methods or inconsistently defined pur-
scribed the purpose of their qualitative studies in poses across the various sections of the research
the following ways: report. Stem (1978) described her entire research
methodology simply as “constant comparative
In this paper I shall discuss a research ap- analysis” @. 50). Kueffner (1976) concluded her
proach . . . in which a given problem area is report with statements reflecting a purpose of
studied for the purpose of developing a con- sensitization rather than conceptual development
ceptual framework [emphasis ours] for under- as presented in earlier sections of the article.
standing and explaining what takes place in the
realm of patient care, given certain sets of cir-
cumstances. (Quint, 1967, p. 109) COMMUNICATING QUALITATIVE RESEARCH
The final goal of which the nurse ethnographer One might expect this diversity of purpose
should never lose sight is to grasp the patients across qualitative studies reflected in a diversity
point of view [emphasis ours], relation to life, of final products. A researcher whose goal is to
to his vision of the phenomenon of health and provide baseline data for future research will for-
illness. (Ragucci, 1972, pp. 489-490) mulate a different final report than a researcher
whose aim is to provide a descriptive base for
Quint emphasized the usefulness of qualitative nursing practice. Similarly, the end product of a
research in the generation of theory or the con- grounded theory study will differ from that of a
ceptual rendering of an area of interest. She noted descriptive study. The reader should evaluate the
that qualitative studies often provide the empir- quality and usefulness of a qualitative study within
ical grounding for more rigorously structured re- the context of the author’s purpose. The reader
search. On the other hand, Ragucci, paraphrasing should not expect theory when the purpose has
her previous quote of Malinowski, stressed that been to describe; similarly, a grounded theory
the accurate and detailed description of a point report may not result in recommendations for
of view, a social world, is the major contribution subsequent quantitative study. Authors may use
of qualitative research. We would add that qual- identical data gathering techniques, but with dif-
itative research also serves as an adjunct to stud- ferent ends in mind. A clearly stated purpose will
ies which are primarily quantitative in nature. help the reader formulate a realistic set of expec-
Depending on the researcher’s purpose, qual- tations.
itative methods may be used for instrument de- Existing guidelines for reporting the process
velopment, illustration, sensitization, or concep- and product of qualitative research usually fail to
tualization. In the first instance, qualitative data discuss the issue of purpose. Most are written
are important in developing or refining the in- from the assumption that the author’s goal is to
strument(s) used in a quantitative study. In the generate theory. In fact, many qualitative re-
second instance, limited qualitative data are col- searchers equate excellence in qualitative reports
lected and used to illustrate the results of a larger, with going beyond description to conceptualiza-
quantitative project. As a sensitizing device, tion of the data (Becker, 1970; Glaser & Strauss,
qualitative findings are important in and of them- 1967; Lofland, 1976; Quint, 1967; Schatzman &
selves since it is the richness and detail of the Strauss, 1973). However, the fact remains that
data that give the reader an understanding of the not all qualitative studies result in conceptuali-
subject’s social world. As the raw material of zations because not all qualitative researchers have
theory, qualitative data are important as a means that purpose in mind.
to an end. The raw data are translated into con- The remainder of this article delineates how
cepts and, in turn, used to illustrate the concepts. reports of qualitative research with differing pur-
In this case, the investigator uses the raw data poses are likely to vary when reported in the
primarily as a catalyst for conceptualization. standardized format of several nursing research
In reviewing published qualitative studies, the journals. Table 1 is a guideline for either eval-
reporting styles reflected an absence of guidelines uating or reporting the results of a qualitative
Table 1. Summary of Guidelines for Reporting Qualitative Research Based on Study Purpose

Parts of Manuscript
Research Purpose Introduction Sample Measures Procedure ResuIts Discussion

Instrumentation Emphasis on final Comparable to Description of Qualitative aspect Qualitative aspect Not mentioned in
project and how subjects in final nature and of study not not mentioned discussion unless
qualitative work project development of mentioned qualitative data
will contribute to interview guide or linked to specific
instrument observer‘s role. methodological or
development Discussion of how research
qualitative data im plications
were collected, ?
categorized, and e
used in
instrument
development
W
rn

I I I ustration Emphasis on Com parable to Description of Brief description Interspersing of Not mentioned i n K
research problem subjects in larger nature and of how qualitative qualitative data to discussion unless z
0
being addressed project or sub- development of aspect of study highlight qualitative data I
in larger project sample from interview guide or was completed qua nti to tive linked to specific
.
x
larger project observer‘s role findings methodological or Z
research R
r

im pl icat ions D
Z
U
Sensitization Emphasis on lack Representative of Description of Explanation of Grouping of data Summary of I
of research and/or target group or nature and how and why into categories or maior themes and 0
4
need for indepth organization development of data collection types which reflect identification or 2
understanding of interview guide or techniques lead to subjects’ views practice and U

subject observer’s role an indepth research


description of implications
group or
organization
Conceptualization Emphasis on lack Theoretical Description of Demonstration of Translation of raw Summary of
(Theory Building) of theory in a nature and how emergent data into theory or major
given substantive development of theory guided theoretical concepts and
area interview guide or data collection formulations identification of
observer’s role illustrated by the practice and
data research
implications
20 RESEARCH IN NURSING AND HEALTH

project when reported in the format adopted by certainty and transformed to a 54-item Likefi scale
the American Psychological Association (Publi- which was administered to a larger sample of
cation Manual of the American Psychological hospitalized patients and then factor analyzed.
Association, 1983). We suggest that qualitative The entire report of the qualitative component
researchers explicitly structure their final reports was written in two paragraphs, which succinctly
in terms of their original purpose. Similarly, we and adequately presented this aspect of the study.
advise the reader to formulate expectations and When qualitative data are used for the purpose
judgments of the work based on the author’s re- of instrumentation, there is no necd to refer to
search purpose. this aspect of the study in the procedure subsec-
tion of the report, since it has been described
lnsfrumenfafion and Illustration
sufficiently in the measures subsection. How-
ever, if the purpose is illustration, the author should
When the investigator is using qualitative data indicate in the procedure subsection how the
to develop a structured instrument or to illustrate qualitative data were collected (i.e. , interviews
quantitative results, the qualitative component is or participant observation).
usually a small piece of a larger quantitative study. When qualitative data are collected for instru-
This aspect of the study should receive compar- ment development, the data are not mentioned in
atively little attention in the final reporting of the results section of the report; rather, as indi-
results. In the beginning section of the report, the cated above, are described in the measures sub-
reader is introduced only to the questions and section. If the qualitative data are collected to
purposes of the larger quantitative project. The illustrate quantitative findings, they are inter-
qualitative aspect of the study is described briefly spersed sparingly throughout the results section.
in the methods section of the article. Nonproba- For example, h e r (1968) interviewed47 newly
bility or convenience sampling for the qualitative employed graduates of three baccalaureate nurs-
component of the study often is the sampling ing programs to supplement quantitative data ob-
method used. The reader should expect the sub- tained on the nurses’ service, professional, and
jects to be comparable to those of the larger study. bureaucratic role orientations in their new jobs.
In the measures subsection, the instrument(s) Brief excerpts from interview transcripts were re-
used to complete the qualitative aspect of the study ported in the results section to illustrate or clarify
is described. Unstructured or semistructured (also the meaning of group scores from quantitative
referred to as intensive or indepth) interviewing measurements. If the author identifies specific
and participant observation frequently are used methodological or researcli implications based on
as data collection techniques. When interviews the qualitative data then this aspect of the study
have been conducted, the author should indicate is included in the discussion section of the report.
the sources of questions for the interview guide However, the author should not include specific
and the major topical areas covered in the guide. qualitative findings in the discussion unless these
If the guide(s) has been pretested, the results of already have been introduced in the measures or
such testing in terms of revision(s) of the instru- results subsections of the paper.
ment should be noted. If data have been collected
using participant observation, then the research-
Sensitization
er’s role in relation to the subjects should be de-
scribed. The description should include the na- When the researcher is reporting the results of
ture and extent of the researcher’s participation a descriptive study, the purpose of which is to
in the organization or group studied and how the sensitize the reader to the viewpoint of a partic-
role developed and changed over the course of ular group, the introduction should show that either
data collection. few (if any) studies have been done on the topic,
If the qualitative data are used as a basis for or those that have been done have failed to rep-
instrumentation, the measures subsection also resent the groups’ point of view.
should describe how qualitative data were cate- Since the author’s purpose is to represent ac-
gorized and used to develop a structured instru- curately and fully the perspective of a particular
ment. For example, in developing an instrument group, the author should specify, in describing
to measure uncertainty in illness, Mishel (198 1) the sample, the characteristics of the population
initially interviewed 45 hospitalized patients to of which the subjects are representative. The
identify statements of uncertainty associated with measures subsection should be similar in content
illness or hospitalization. The qualitative data were to that described in the preceding section of this
categorized according to sources and types of un- paper. The development and pretesting of inter-
QUALITATIVE RESEARCH I KNAFL AND HOWARD 21

view guides andfor the planning for and the ac- combined. In general, the results section should
tual development of the researcher’s role vis-a- comprise a synthesis of the subjects’ viewpoints,
vis the subjects should be presented. either in general or with regard to the specific
In the procedure subsection, the author should topic studied. The presentation must be selective
describe carefully and succinctly how the data as the investigator inevitably has more data than
collection and analysis techniques were carried reasonably can be present in one report.
out to insure that the reader will understand how The specific organizing framework for pre-
the approach fulfilled the original research pur- senting the analysis of data will vary consider-
pose. Varying approaches to reporting the re- ably from study to study. For example, Ipema
search procedure are suggested in textbooks on (1979) studied the experience of and recovery
qualitative methods. Bogdan and Taylor (1975) from rape by interviewing 11 rape victims on two
listed seven procedure topics: method, time and separate occasions. From transcripts of the tape
length of study, nature and number of settings recorded interviews, three major content cate-
and subjects, how the subject became a subject, gories were identified: the victim’s report of rape,
the researcher’s frame of mind, researcher-sub- rape sequelae, and disruptions of the social sys-
ject relationship, and checks on the data (p. 143). tem. In the results section, Ipema chose to present
Lofland (1971) listed five topics: inception and the category of “rape sequelae” more fully than
social relations, personal feelings, materials col- the other two categories, using both direct quotes
lection, analysis, and retrospect (p. 131). The from interviews and repetition of subjects’ sim-
difference between the two sets of guidelines ilar responses to convey the rape victims’ com-
highlights the lack of consensus among qualita- mon experiences.
tive researchers regarding how their work should Hampe (1975), on the other hand, used liter-
be reported. ature on needs of the grieving person to identify
Although differing in content, guidelines are major topics for two interview schedules. Data
useful to the extent that they provide the writer from two tape recorded interviews with 27 spouses
with an organizing framework for presenting the of terminally ill patients were then presented in
study design. The main problem with such guide- the results section to demonstrate the meaning of
lines is a practical one having to do with space each category of need to the subjects. The pur-
constraints confronting any author who writes for pose of sensitization is found in the author’s fol-
publication in a journal. Lofland (1971), for ex- lowing statement introducing verbatim excerpts
ample, listed 39 questions or topics to be ad- of data presented with some of the category de-
dressed under five broad headings. scriptions: “Because the impact of the words of
We suggest the following as minimal require- the spouses was so strong and at times heartrend-
ments for what should be reported in the proce- ing, the significance of the study can be most
dure subsection of a qualitative report: clearly exemplified when the spouses’ comments
speak for themselves” (p. 116).
1. Preparation for data collection, including
The final discussion section in a descriptive
gaining access to study sites and subjects
report should focus on implications for practice
and training of project staff.
2. Length of time spent collecting data, how and future research. Specific implications are
identified and discussed with reference to the or-
data were recorded, and the amount of data
ganizing framework and major themes presented
collected.
3. Steps taken to organize, categorize, or in the results section of the paper. If the author
summarize the data prior to final analysis. has fulfilled the purpose stated at the outset, then
4. Management of threats to the validity and the reader is sensitized to the perspective of the
group being studied. In the final section, the
reliability of the data.
nursing implications of the perspectives de-
5. Process by which conclusions were derived
scribed are presented. The author also may sug-
from the data.
gest avenues for additional research and prom-
Depending on space limitations and the nature of :sing approaches for such research.
the study being reported, the author may decide
to devote comparatively more or less attention to
Conceptualization
any of the preceding topics or to add topics par-
ticularly relevant to a study. Several nurse authors support an inductive ap-
More so than in reports of experimental and proach to theory development in nursing practice
survey research, in the results section of a qual- through the collection and use of qualitative data
itative article, data reporting and discussion are (Jackson, 1975; Jacobsen, 1970; Quint, 1967;
22 RESEARCH IN NURSING AND HEALTH

Wilson, 1977). As Jacobsen (1970) noted, As stated previously, the measures subsection
“Nursing is in need of substantive theories ap- should describe the development and nature of
plicable in a variety of nursing situations. It is data collection instruments. When the author’s
possible to develop such theories directly from purpose is conceptualization, changes in the in-
the qualitative data of everyday nursing con- struments during the course of data collection
texts” (p. 13). should be noted and justified in terms of their
Investigators reporting such studies should in- theoretical relevance. As in the sampling subsec-
troduce the reader to the conceptual or theoretical tion, the author needs to make explicit the fit
significance of the subject matter. For example, between ongoing methodological decisions and
Wilson (1977) placed her study of an experimen- the developing theory.
tal treatment community within the context of In a conceptualization study, as with previ-
historical social rejection of the insane. The con- ously discussed descriptive studies, the proce-
ceptual significance of her study, implied in the dure subsection should contain what specific data
introductory information, explained how a non- collection techniques were used, the time span
traditional psychiatric treatment center survives over which data was collected, the quantity of
in the midst of a potentially rejecting public. When data collected, steps taken to organize and process
the purpose is to develop theory, it is important the data, and limitations of the data. The author
to alert the reader from the outset that the report should demonstrate that specific data collection
is aimed at conceptualization. The study ap- and processing steps served the overall theoreti-
proach may be introduced as an extension of ex- cal purpose.
isting descriptive research, with an emphasis on Research by Kueffner (1976) illustrates this
identifying conceptual links among categories of point. In initiating a study of hospitalized chil-
data rather than simply describing the data to dren on isolation technique, the investigator se-
communicate thc subject’s point of view. This lected a sample of children with severe burns,
being the case, it is necessary to review other because “they are inevitably isolated, generally
theories pertinent to the subject matter and to for extended periods” (p. 183). However, the
discuss the rationale for additional theorizing. author commented on how the original theoreti-
Studies often are described explicitly in terms cal notion was altered by her observations as data
of prhciples of grounded theory (Glaser & Strauss, gathering progressed. “Within the research de-
1967). Integral to this theory-discovering ap- sign, it became increasingly difficult to identify
proach is theoretical sampling. The technique behaviors relating solely to the isolation experi-
precludes specifying one’s ‘entire sampling de- ence. It was the burn with its all-encompassing
sign prior to data collection. Sampling decisions pain that dominated the situation” (p. 185). The
are dependent on analysis of the incoming data author later incorporated this alteration in con-
and the developing theory. The author should ceptualization with the original concept of iso-
present major sampling decisions made during lation and a new category of data, “loss,” in her
the course of data collection. These usually entail formulation of a theoretical model of the passage
changes in the settings or subjects being studied through hospitalization of a severely burned school
or in the timing of observations. Further, the au- age child. In describing her model, Kueffner
thor should state how changes were justified in stated, “The pain, sense of aloneness, and feel-
terms of the developing theory. For example, a ings of loss experienced during hospitalization
researcher who is conceptualizing the relation- and stemming from the bum and the state of iso-
ship between staff nurses and patients’ relatives lation, exert the dominant influence on behavior”
based on observations made on adult acute care (P. IN).
units may choose to refine and further expand the In the results section, the author’s conceptual
developing theory by collecting data on pediatric rendering of the data is presented. Specific con-
units or on units where a different mode of nurs- cepts or categories of data are discussed in terms
ing care is practiced. Such changes and the ra- of their place within and contribution to the au-
tionales underlying them should be summarized thor’s overall conceptualization of the phenom-
in the sampling subsection. Because this is a non- enon under study. Lofland (1971) said that
conventional sampling approach, it is advisable “penetrating and useful qualitative analysis has
to cite articles or texts which address the appro- the feature of striking a balance between abstract
priateness and use of theoretical sampling for the and general concepts on the one hand and de-
purpose of theory development (Glaser & Strauss, scription and quotations from a settings’ partici-
1967; Wilson, 1977). pants on the other” @, 128). This balance is
QUALITATIVE RESEARCH / KNAFL AND HOWARD 23

achieved test by interweaving theory and exam- poses. These varying purposes are, in turn, re-
ples of the data from which the theory was de- flected in research reports with differing em-
rived. phases and structures. This is especially true with
Presenting and discussing specific results in the regard to sensitization and conceptualization.
context of their theoretical relevance demon- Descriptive reports too often are criticized for not
strates how conceptual formulations are grounded going beyond description; grounded theories are
in the data. The author should indicate the nature criticized for not presenting detailed descriptions
and scope of the data underlying conclusions and of settings and subjects. Such criticisms are un-
highlight the report with representative quotes from justified when viewed in terms of study purpose.
the data. Lofland (1974) suggested that approx- They reflect misunderstanding on the part of the
imately 60% of the report be devoted to concep- reader, inadequate reporting on the part of the
tualization and 40% to presentation of data. While researcher, or both. With these guidelines we have
these percentages are likely to vary across arti- sought to reduce both misunderstanding and poor
cles, it is reasonable to expect that a grounded reporting. We ask readers of qualitative research
theory report should have proportionately more to evaluate that work in the context of overall
theory than description. study purpose. We ask writers to state their pur-
Lofland (1974) described the blending of re- pose explicitly so that the reader can formulate
sults, “frame” (conceptualization), and discus- realistic expectations.
sion: As we have shown, varying reports of quali-
tative research can be adapted to a standard jour-
Frame and qualitative materials coexist as one nal format. However, one might expect reports
whole, each depending upon the other for the of qualitative studies to continue to emphasize
“interest” a reader has in the frame or in the different aspects of the research or indicate that
qualitative material . . . The frame taken sep- the research was carried out in differing ways.
arately is dull because the reader has little con- This is as it should be since qualitative studies
ception of the concrete empirical reality to which serve a variety of research purposes, which re-
the frame might refer. The “data” alone are quire varying research designs and result in sev-
dull because the reader has no notion of what eral styles of research reporting.
sort of social structure or process might be in-
volved. But interpenetrated through minute and
continual alternation between data and frame-
R€F€R€NC€S
\
elements, the whole is more than the part. (pp.
108-109) Becker, H. S. Sociologim/ work. Chicago: Aldine, 1970.
Bogdan, R., & Taylor, S. Introduction to qualitative
As in virtually all reports of nursing research, research methods: A phenomenological apprwch to
the author should conclude the final discussion the social sciences. New York: Wiley, 1975.
section with statements of the implications for Cook, T. D., & Reichordt, C. S. Qualitative and quan-
nursing practice and further research and theory titative methods in evaluotion research. Beverly Hills:
Sage, 1979.
development needed in the area. Such implica-
Glaser, B., & Strauss, A. The purpose and credibility
tions and recommendations should be linked ex- of qualitative research. Nursing Research, 1966,
plicitly to the author’s conceptual formulations. 15, 56-61.
Recognizing the fact that nursing as a scientific Glaser, B., & Strauss, A. The discovery of grounded
discipline has a relatively short history of pub- theory: Strategies for quolitotive research. Chicago:
lished research endeavors, we felt that clarifying Aldine, 1967.
the purposes of qualitative research would benefit Hampe, S. 0. Needs of the grieving spouses in a
investigators who need to know and understand hospital setting. Nursing Reseorch, 1975, 24,
a variety of research methodologies. We partic- 1 13-1 20.
ularly wanted to suggest guidelines for reporting Ipema, D. K. Rape: The process of recovery. Nursing
Research, 1979, 28, 272-275.
qualitative, studies, because we believe, as do
Jackson, 8. S. An experience in participant obser-
others (Glaser & Strauss, 1966; Klenow, 1981; vation. Nursing Outlook, 1975, 23, 552-555.
Quint, 1967), that existing misunderstandings re- Jacobsen, M. Qualitotive data as a potential source
late to the lack of standardization in reporting this of theory in nursing. Image, 1979, 4 , 19-14.
type of research. Klenow, D. J. Qualitative methodology: A neglected
The data collection techniques associated with resource in nursing research. Research in Nursing
qualitative studies serve a variety of research pur- and HeoIth, 1981, 4 , 281-282.
24 RESEARCH IN NURSING AND HEALTH

Kramer, M. Role models, role conceptions, and role Publication manual of the American Psychological As-
deprivation. Nursing Reseorch, 1968, 77, 1 15-1 20. sociation (3rd ed.). Washington, D.C.: American
Kramer, M. Collegiate graduate nurses in medical Psychological Association, 1983.
center hospitals. Nursing Research, 1969, 18, Quint, J. C. The case for theories generated from
196-220. empirical data. Nursing Research, 1967, 16,
Kueffner, M. Passage through hospitalization of se- 109-1 14.
verely burned, isolated school-age children. Com- Ragucci, A. T. The ethnographic approach and nurs-
municating Nursing Research, 1976, 7, 1 8 1-1 97. ing research. Nursing Research, 1972,21, 485-490.
Lofland, J. Analyzing social settings: A guide to qual- Schatzman, L. & Strauss, A. Field research: Strategies
itative observation and analysis. Belmont, CA: for natuml sociology. Englewood Cliffs, NJ: Pren-
Wadsworth, 1971. tice Hall, 1973.
Lofland, J. Styles of reporting qualitative field re- Stern, P. N. Stepfather families: Integration around
search. The American Sociologist, 1974,9, 101-1 1 1 . child discipline. Issues in Mental Health Nursing,
Lofland, J. Doing social life: The qualitative study of 1978, I , 50-56.
human interaction in natural settings. New York: Wilson, H. A. Limiting i n t r u s i o 4 o c i a l control of
Wiley, 1976. outsiders in a healing community. Nursing Re-
Mishel, M. H. The measurement of uncertainty in search, 1977,26, 103-1 1 1.
illness. Nursing Research, 1981, 30, 258-263.