The Quality of Qualitative Research

Dave S. Collingridge, PhD

Edwin E. Gantt, PhD

In general, an appreciation of the standards of qualitative research, its popularity will continue to
qualitative research and the types of qualitative grow in the health sciences.
data analyses available to researchers have not kept Notwithstanding the growing popularity of
pace with the growing presence of qualitative stud- qualitative methods, knowledge of the standards of
ies in medical science. To help rectify this problem, qualitative research and the specific information
the authors clarify qualitative research reliability, that it is capable of providing is lagging. Consider
validity, sampling, and generalizability. They also that within medicine, qualitative studies are often
provide 3 major theoretical frameworks for data col- relegated to the soft domains of science, and many
lection and analysis that investigators may consider are reluctant to accept the validity and usefulness
adopting. These 3 approaches are ethnography, exis- of qualitative research.4 Added to this problem are
tential phenomenology, and grounded theory. For poorly executed qualitative studies that have made
each, the basic steps of data collection and analysis their way into the health care literature,6,7 and
involved are presented, along with real-life examples other qualitative studies where the authors failed
of how they can contribute to improving medical to report the approaches they used to analyze qual-
care. (Am J Med Qual 2008;23:389-395) itative data.8,9 These issues raise concerns about
Keywords: qualitative; quantitative; validity; reliability; clinicians’ current levels of understanding regard-
sampling; phenomenology; ethnography; grounded theory ing the proper application of qualitative methods
in medical research.
In recent years qualitative research has been rec- Clearly, further discussion of the standards of
ognized as making a valuable contribution to med- qualitative research and the types of data analysis
ical science.1,2,3 The growing popularity of approaches available to researchers is needed. This
qualitative research is fueled in part by the real- article attempts to remedy these problems by dis-
ization that traditional quantitative research is cussing fundamental issues related to rigorous
limited in its ability to capture the meanings qualitative medical research. These issues are reli-
people attach to health care social phenomena, and ability, validity, sampling, and generalizability. We
that understanding the experiential and interpre- compare the qualitative and traditional quantita-
tive elements of medical practice that qualitative tive understandings of these concepts to clear up
research is capable of providing is essential to the confusion that currently surrounds the issue of
enhancing clinical knowledge and care.4,5 Certainly rigor in qualitative studies. Next we discuss spe-
as more people recognize the unique benefits of cific qualitative data analysis approaches, how
they should be implemented, and what they can
contribute to medical science. We introduce the
AUTHORS’ NOTE: Dr Collingridge is with Intermountain reader to 3 major qualitative approaches: (a) ethnog-
Healthcare, Salt Lake City, Utah. Dr Gantt is with Brigham raphy, which reveals the behaviors and attitudes of
Young University, Provo, Utah. The authors disclosed that they a cultural group; (b) existential phenomenology,
have no financial and personal relationships that inappropriately
which uncovers the experiential essence of phe-
influenced their actions and created a conflict of interest.
Corresponding author: Dave Collingridge, PhD, Intermountain nomena; and (c) grounded theory, the purpose
Healthcare, 8th Avenue and C Street, Salt Lake City, UT 84143 of which is to build theories of social phenomena.
(email: david. [email protected]). We outline the basic steps involved in these
American Journal of Medical Quality, Vol. 23, No. 5, Sep/Oct 2008 approaches and provide examples to illustrate the
DOI: 10.1177/1062860608320646 contributions they are capable of making to med-
Copyright © 2008 by the American College of Medical Quality ical knowledge.

389
390 Collingridge, Gantt AMERICAN JOURNAL OF MEDICAL QUALITY

SCIENTIFIC STANDARDS This concept of reliability differs from the traditional

quantitative understanding in that the focus is not
Influential philosopher of science, Thomas Kuhn, on obtaining exactly the same results time and
stated that it is difficult to evaluate the merits of again, but rather on achieving consistent similarity
one research paradigm from the perspective of a dif- in the quality of the results.
ferent paradigm because of different standards of The difference between quantitative and qualita-
research.10 Such is the case with qualitative and tive reliability is represented in the following exam-
quantitative research. Each has its own standards ple from a study of human consciousness. From a
of scientific research, thus making it difficult to quantitative neurocognitive perspective, reliability
judge the merits of one using the criteria of the in a study of human consciousness may be uncover-
other. These differences in standards often create ing the processes and elements of conscious experi-
difficulties for medical professionals who must judge ence that are the same for a homogenous group of
the merits of qualitative research.11 These difficul- people, regardless of time and setting.13 From a
ties may be alleviated if professionals more fully qualitative perspective, reliability is uncovering the
understand the differences between qualitative and rich meanings inherent in people’s conscious expe-
quantitative standards of scientific rigor that follow. riences.14 When those meanings differ, it is not
because an unreliable method was used; rather, dif-
Reliability ferences in meaning stem from the fact that people
experience social phenomena in different ways.
Replication of research findings is important in It is important to note that the qualitative con-
traditional quantitative research. It facilitates the cept of reliability does not minimize the importance
scientific self-correction process whereby ideas are of finding similar results given similar research set-
modified or replaced as science comes to a more accu- tings. Two researchers who employ the same quali-
rate understanding of the natural world. Replication tative method in similar settings with similar
of research findings increases our confidence in the people should get similar results. However, if the
accuracy of hypotheses, whereas a lack of replication results are not exactly the same, this does not mean
decreases our confidence in hypotheses. Central to that the qualitative method employed is unreliable.
this process is using reliable research methods. It may be that both researchers uncovered slightly
According to the quantitative research paradigm, different, but legitimate, understandings of the
our methods are considered reliable if they are suffi- same phenomenon. Or it may be that one researcher
ciently free of bias to consistently produce the same achieved a more insightful and nuanced level of
results given similar contexts (ie, similar partici- understanding than the other. Slight differences in
pants, similar research conditions). Methods that are findings should not immediately be construed as the
reliable in the sense of being free from bias are result of employing a flawed method of investigation
thought to provide us with an accurate representa- or analysis. Differences often reflect a multifaceted
tion of the natural world. Conversely, methods that understanding of complex social phenomena.2 When
lack reliability are not believed to provide us with an differences arise, additional investigations may be
accurate representation of the world. needed to achieve a more complete understanding of
Reliability is an equally important issue in quali- the phenomenon of interest.
tative research. However, what constitutes a reliable
finding is often understood in a slightly different
way. Reliability in qualitative research typically Validity
refers to adopting research methods that are
accepted by the research community as legitimate In general terms, validity refers to the extent to
ways of collecting and analyzing data. Specifically, which research measures what it purports to meas-
reliable qualitative methods consistently produce ure. In quantitative research, concern with validity
rich and meaningful descriptions of phenomena. leads us to ask questions such as: “Can we identify
Thus qualitative researchers who adopt reliable, the causes of a rare disease using a retrospective
qualitative methods and conduct their analyses in a case control study design?” “Does the Mini Mental
competent manner (see validity) are expected to pro- State Exam provide an adequate measure of cogni-
duce results that enrich our understanding of the tive sequelae?” and “Is magnetic resonance imaging
meanings that people attach to social phenomena.12 an appropriate measure of cognitive functionality?”
AMERICAN JOURNAL OF MEDICAL QUALITY Qualitative Research Quality 391

Three subcategories of validity are construct valid- Participant sampling in qualitative research is
ity, content validity, and criterion validity. equally rigorous (ie, appropriate to the task at hand
Construct validity reflects a concern with whether and the research questions being asked), but in
we are truly assessing an underlying construct ways that differ from the quantitative paradigm.
such as emotional well-being, physical comfort, and Qualitative research participant sampling should
pain. Content validity reflects a concern with follow a well-defined rationale16 and fulfill a specific
whether our measurement tools provide an ade- purpose, which is why qualitative sampling is often
quate measure of the construct of interest. Lastly, called purposive. In particular, purposive sampling
criterion validity refers to the strength of the rela- refers to selecting participants who serve a specific
tionship between our measurement tools and other purpose consistent with a study’s main objective.
measures of the same phenomenon. There are several types of purposive sampling
Validity is construed in a similar fashion in quali- techniques. For instance, criterion sampling, which
tative research. Measuring what one purports to is selecting participants who have experienced
measure in qualitative research means selecting an the phenomenon of interest, is particularly impor-
appropriate method for a given question and apply- tant in phenomenology where the goal is to investi-
ing that method in a coherent, justifiable, and rigor- gate experiential phenomenon. Phenomenological
ous manner. Careful craftsmanship and attention to researchers may also use maximum variation sam-
detail makes legitimate results possible and allows pling, which involves selecting participants who can
for defensible knowledge claims.15 For instance, phe- provide a variety of descriptions of the phenomenon
nomenological researchers who investigate experien- of interest. A purposive technique that may be used
tial phenomena must make sure that they facilitate in grounded theory is theoretical sampling.
a free-flowing and intensive discussion about events Theoretical sampling in grounded theory involves
with participants to elicit experientially vibrant selecting participants who are capable of providing
responses. Grounded theory researchers must make insight into emerging theoretical constructs. On the
sure that they are asking the right questions of the other hand, an ethnographer may rely on oppor-
right people to uncover emerging theoretical cate- tunistic sampling. Ethnographic opportunistic sam-
gories. Ethnographic researchers must be certain pling involves seeking out opportunities to talk with
that they are observing people and events in their and observe people who illuminate the researcher’s
proper settings to ensure that their cultural descrip- understanding of a culture or group. Although these
tions of groups are accurate and concrete. sampling techniques differ considerably from tradi-
Thus we see that qualitative and quantitative inter- tional random sampling techniques, they are
pretations of validity are similar. Both paradigms are equally rigorous in their application.
equally concerned with producing legitimate results.
Qualitative research involves (a) accurately describing Generalizability
intangible constructs such as experience, theory, and
culture (ie, construct validity); (b) adequately accessing Extending the results of a given study to a
constructs through effective interviewing and observa- broader group of people is important in medical sci-
tion techniques (ie, content validity); and (c) obtaining ence where there is an ongoing concern with
results that agree with other measures of the same improving the health care status of people. The
phenomena (ie, criterion validity). question as to whether qualitative results are gen-
eralizable in the same way that quantitative results
Sampling are generalizable is a matter of ongoing debate.15
Some may argue that because purposive sampling is
Participant sampling techniques in the quantita- not random, we cannot justify generalizations based
tive paradigm typically aim to achieve a degree of on a random sampling model. Although this is true,
random selection. The main reason for random sam- it is important to note that generalizability is not
pling is to enhance the generalizability of the limited to probability sampling theory. There are dif-
results, or the extent to which the results can be ferent ways of understanding generalizability.
applied to people and contexts outside of the study. An alternative interpretation of generalizabil-
If we randomly select participants, then we are jus- ity, known as analytical generalization, involves
tified, to a certain degree, in referring our results making a “reasoned judgment about the extent
back to the population from where the sample came. to which the findings in one study can be used as
392 Collingridge, Gantt AMERICAN JOURNAL OF MEDICAL QUALITY

a guide to what might occur in another situa- Qualitative techniques refer to modes of data col-
tion.” 17(pp231–235) Analytical generalization relies on lection in qualitative research, such as compiling
assertational logic whereby researchers point out written responses, interviewing, and conducting
similarities (and differences) between situations focus groups. Such techniques are sometimes
and draw on relevant theoretical frameworks for used in conjunction with other data-gathering
interpretive understanding to support their gener- techniques, as in the case of requesting open-ended
alization claims. This process allows others to written responses at the end of a Likert-style ques-
decide the soundness of the arguments being made tionnaire. In such cases we may talk of using a
and the applicability of the generalizations offered. qualitative technique, but not necessarily a major
An example of the analytical generalization qualitative approach. Following a qualitative
approach is the proximal similarity model.18 approach involves much more than just using qual-
According to this model, generalizability depends on itative data-gathering techniques. It involves select-
the extent to which a study’s context is similar to the ing a theoretical framework for answering a specific
natural context of the phenomenon in question. research question and employing data-gathering
Similarities in context are evaluated using the and analysis procedures consistent with that frame-
model’s gradient of similarity criterion, which work. Three major frameworks that we will consider
involves assessing the degree of similarity between are ethnography, existential phenomenology, and
the time frame, place, people, and setting for both the grounded theory.
study and outside groups. If the contexts are too dis-
similar, the generalization is not warranted. If the 2 Ethnography
contexts are similar, then we can generalize with a
certain degree of confidence.19 This approach has The purpose of ethnography is to understand
much in common with the study of case law, where the behaviors and attitudes of a cultural group.
applicable precedent and generalizability is estab- Ethnographers seek this understanding by gaining
lished through the analysis of the specific character- a native perspective from the inside looking
istics and resemblances of particular cases.17(pp233–234) around, rather than from the outside looking in.21
There are other criteria that should be taken into A native perspective is usually obtained by enter-
consideration when judging the generalizability of ing the group’s environment to interview people
a qualitative study. Qualitative studies that do not and observe objects, events, and symbols that
follow a coherent qualitative framework and are define the culture and give it meaning. Occasionally,
poorly executed lack generalizable results. Also, where appropriate, researchers may participate in
single case studies that provide insights into 1 per- noteworthy communal events.
son and descriptive studies that are limited to Most well-constructed ethnographic studies gen-
descriptions of phenomena lack generalizability. erally follow the following format of data collection
However, studies that build on existing theoretical and analysis: (a) collecting data through natural
concepts through comprehensive literature reviews, observation; (b) creating a textual representation
employ theory-based sampling procedures, follow of the data; (c) reading through the text and form-
well-defined data analysis procedures, clearly ing initial codes; (d) analyzing the codes to identify
define how the findings apply to other contexts, and themes and patterns; (e) interpreting the identified
integrate results into existing research in a coher- themes and patterns; and (f) creating a narrative
ent fashion produce results that can be generalized account that describes the people, places, and
with a greater degree of confidence.20 important objects and social events that define the
culture and give it meaning.16
THREE MAJOR QUALITATIVE APPROACHES Ethnographic research can benefit health care
research in many ways. A well-designed ethno-
As pointed out earlier, a barrier to qualitative graphic study that builds on existing research and
research gaining more respect in medical science is utilizes theory-based sampling strategies has the
a lack of familiarity with major approaches. Before potential to illuminate our understanding of rela-
considering 3 qualitative approaches, it is impor- tional phenomena that affect the quality of health
tant to address the often confusing distinction care. For example, ethnographic research may
between qualitative techniques and approaches. improve our understanding of the nurse–physician
AMERICAN JOURNAL OF MEDICAL QUALITY Qualitative Research Quality 393

working relationship, the communication culture Existential phenomenological research can bene-
among critical care providers, the culture of crew fit health care science in several ways. It is particu-
resource management among emergency staff, and larly suited for identifying the foundational aspects
the dynamics of the provider–family relationship of what is taking place in a health care setting from
in family medicine. the perspective of those who provide and receive the
For example, take Richardson’s study investigat- care. For instance, it can reveal the essence of
ing the culture of physiotherapy practice and how it employee engagement, uncover the fundamental
influences professional development in graduate stu- nature of the patient–provider relationship, reveal
dents.22 She observed physiotherapy sessions to physicians’ experiences with new technologies, and
learn how practitioners’ workplace activities and uncover subtle workplace circumstances that hinder
attitudes influenced each other. She also interviewed nurses from fulfilling their primary responsibility to
therapists to better understand their perspectives heal. Without first uncovering the experiential foun-
and experiences. Analysis of the data elicited valu- dations of human phenomena, our ability to effect
able information on how therapists’ attitudes vary positive change may be limited.25 For this reason,
by area of practice, and how these attitudes influence existential phenomenological research is a good
students’ perceptions of physiotherapy. The results starting point when seeking to improve medical care.
led her to conclude that students need to have For example, Rieman investigated the essential
strong, favorable attitudes toward physiotherapy structure of the nurse–patient care interaction
against which they can evaluate their workplace using phenomenology.26 In an interview setting,
experiences before entering the workforce as practi- Rieman captured patients’ descriptions of
tioners. This study has important implications for nurse–patient interactions on audiotape. The
the education and training of health care providers. interviews were then transcribed and analyzed to
create in-depth descriptions of 2 interactions: the
Existential Phenomenology caring nurse interaction and the noncaring nurse
interaction. Results showed that the caring inter-
The purpose of phenomenology is to understand action is not only defined by what the nurses phys-
phenomena from the perspective of those who ically do, but by who they are in the eyes of the
experience the phenomena. Specifically, the aim is patient. Demonstrating genuine interest in the
to know an experience the way that the partici- patient by carefully listening was the most impor-
pants know it, to understand the meanings they tant aspect of the nurse–patient caring relation-
attach to their experiences, and to capture the ship. This study has important implications for
essence of a phenomenon as they experience it.23 nursing theory, training, and practice.
This understanding is primarily achieved through
interpersonal interviews. “Interviews are particu- Grounded Theory
larly suited for studying people’s understandings
of the meanings in their lived world, describing The purpose of grounded theory is to build or
their experiences and self-understanding, and clar- expand on theories about human phenomena.
ifying and elaborating their own perspective on Grounded theory research is unique in the sense
their lived world.”17(p105) that it is an iterative process wherein the grounded
Most well-constructed phenomenological studies theorist continually goes back to the field to collect
generally utilize the following data collection and more data and answer new questions. This process
analysis process: (a) interpersonal interviews with continues until no new information is gleaned dur-
up to 10 people who are willing to share their expe- ing the data gathering process. As the information is
riences, (b) transcribing the interview data, collected and organized “into a logical, systematic,
(c) locating relevant statements in the transcripts and explanatory scheme,” thematic categories begin
that express self-contained units of meaning, to emerge.27 The categories are eventually organized
(d) identifying the meanings contained in each into a comprehensive whole, thus creating a sub-
statement, (e) synthesizing the meaning units into stantive theory grounded in the experiences of those
common themes, and (f) synthesizing the themes familiar with the phenomenon of interest.16
across interviews to create a general description Most grounded theorists generally go along with
of what it is like to experience the phenomenon the following process of data gathering and analysis:
of interest.24 (a) interviews with 20 to 30 people who are familiar
394 Collingridge, Gantt AMERICAN JOURNAL OF MEDICAL QUALITY

with the phenomenon of interest; (b) transcribing and quality of care. However, an understanding of
the interview data; (c) open coding, which involves the standards of rigorous qualitative research and
identifying relevant concepts in the text; (d) con- familiarity with qualitative approaches has not
stantly comparing open codes looking for conceptual kept pace with the growing presence of qualitative
similarities and differences; (e) identifying emerging methods. This disparity has led to some poorly con-
theoretical concepts; (f) continued sampling and structed qualitative studies, confusion over the
interviewing as theoretical categories emerge and merits of qualitative research, and a lack of report-
novel questions arise; (g) continued coding and com- ing on adopted theoretical frameworks and data
parison of codes until nothing new is added to the analysis procedures.
theoretical categories; and (h) synthesizing the the- We have attempted to rectify these problems by
oretical categories in a coherent fashion, including discussing reliability, validity, sampling, and gener-
defining core categories, visually displaying rela- alizability in qualitative research. Also, to promote
tionships between categories, and linking theoreti- the proper analysis of data and encourage the
cal concepts to the existing literature.28 reporting of data analysis procedures in the litera-
Grounded theory may benefit health care ture, we presented 3 major qualitative approaches
research in many ways. For instance, it can be used to data collection and analysis, namely ethnography,
to expand models of how emergency room workers existential phenomenology, and grounded theory. It
triage patients, to build a model of how emergency should be noted that there are other types of quali-
care physicians make decisions to terminate efforts tative approaches; we chose to discuss ethnography,
to save lives, and to model reliability and trust existential phenomenology, and grounded theory
among critical care workers. Such models provide because of their established reputation and appar-
not only a detailed account of phenomena, they ent usefulness in medical science. Interested read-
also elucidate the why or how of phenomena, ers are encouraged to consult the numerous
thereby providing clinicians with valuable infor- resources on qualitative methods available in the
mation that can be used to inform future research health and social sciences.
and shape medical care policies and procedures. We are confident that qualitative methods will
For example, Stanley used grounded theory to gain further respect and recognition in medical sci-
develop a theory of well-being in geriatric popula- ence as investigators more fully appreciate the
tions.29 Stanley’s analysis of in-depth interviews fundamental standards of qualitative research,
with seniors about their perceptions of well-being clearly describe their data analysis techniques, and
produced an interesting model of well-being in old adopt, when appropriate, clear theoretical frame-
age. She discovered that the core category or main works for their investigations.
issue of concern in seniors’ lives is perceived con-
trol. Perceived control means having control over
one’s routine, making choices, and maintaining a REFERENCES
sense of independence. Four subcategories linked to 1. Shaw RL, Booth A, Sutton AJ, et al. Finding qualitative
perceived control are circumstances that challenge research: an evaluation of search strategies. BMC Med Res
one’s equilibrium, the status of personal resources, Methodol. 2004;4:5.
believing “it’s up to me,” and feeling comfortable 2. Giacomini MK, Cook DJ. Users’ guides to the medical litera-
ture: XXIII. Qualitative research in health care A. Are the
and secure. The model also showed that seniors
results of the study valid? Evidence–based medicine work-
often resort to what she called “trading off.” Trading ing group. JAMA. 2000;284:357-362.
off is replacing something with something else 3. Dixon-Woods M, Sutton A, Shaw R, et al. Appraising quali-
when changing life circumstances challenge one’s tative research for inclusion in systematic reviews: a quan-
sense of autonomy. These findings have important titative and qualitative comparison of three methods. J
Health Serv Res Policy. 2007;12:42-47.
implications for geriatric care and thantanology,
4. Malterud K. The art and science of clinical knowledge: evidence
which is the science of death and dying. beyond measure and numbers. Lancet. 2001;358:397-400.
5. Malterud K. Qualitative research: standards, challenges,
and guidelines. Lancet. 2001;358:483-488.
CONCLUSION 6. Jones R. Strength of evidence in qualitative research. J Clin
Epidemiol. 2007;60:321-323.
In the medical field, qualitative research holds 7. Mays N, Pope C. Assessing quality in qualitative research.
great promise for improving clinical knowledge BMJ. 2000;320:50-52.
AMERICAN JOURNAL OF MEDICAL QUALITY Qualitative Research Quality 395

8. Coiera E, Tombs V. Communication behaviors in a hospital 21. Uzzell D, Barnett J. Ethnographic and action research. In:
setting: an observational study. BMJ. 1998;316:673-676. Breakwell GM, Hammon S, Fife-Schaw C, Smith JA, eds.
9. Pronovost P, Berenholtz S, Dorman T, Lipsett PA, Simmonds Research Methods in Psychology. 3rd ed. Thousand Oaks,
T, Haraden C. Improving communication in the ICU using CA: Sage Publications Inc; 2006:300-320.
daily goals. J Crit Care. 2003;18:71-75. 22. Richardson B. An ethnography of physiotherapy culture. In:
10. Kuhn T. The Structure of Scientific Revolutions. 3rd ed. Finaly L, Ballinger C, eds. Qualitative Research for Allied
Chicago, IL: University of Chicago Press; 1996. Health Professionals: Challenging Choices. West Sussex,
11. Chapple A, Rogers A. Explicit guidelines for qualitative research: England: John Wiley & Sons; 2006.
a step in the right direction, a defense of the ‘soft’ option, or a 23. Spradley J. The Ethnographic Interview. New York: Holt,
form of sociological imperialism? Fam Pract. 1998;15:556-561. Rinehart, & Winston; 1979.
12. Burkett GL, Godkin MA. Qualitative research in family 24. Polkinghorne D. Phenomenological research methods. In:
medicine. J Fam Pract. 1983;16:625-626. Valle SR, Halling S, eds. Existential-Phenomenological
13. Beatty J. The Human Brain: Essentials of Behavioral Perspectives in Psychology. New York: Plenum Press;
Neuroscience. Thousand Oaks, CA: Sage Publications Inc; 2001. 1989.
14. Pollio HR, Henley T, Thompson CB. The Phenomenology of 25. Colaizzi PF. Reflection and Research in Psychology: A
Everyday Life. Cambridge, UK: Cambridge University Press; Phenomenological Study of Learning. Dubuque, IA: Kendall-
1997. Hunt; 1973.
15. Kvale S. Ten standard objections to qualitative research 26. Rieman D. The essential structure of a caring interaction.
interviews. Journal of Phenomenological Psychology. In: Creswell JW, ed. Qualitative Inquiry and Research
1994;25:147-173. Design: Choosing Among the Five Traditions. Thousand
16. Creswell JW. Qualitative Inquiry and Research Design: Oaks, CA: Sage Publications Inc; 1998.
Choosing Among the Five Traditions. Thousand Oaks, CA: 27. Strauss A, Corbin J. Basics of Qualitative Research:
Sage Publications Inc; 1998. Techniques and Procedures for Developing Grounded
17. Kvale S. InterViews: An Introduction to Qualitative Research Theory. Thousand Oaks, CA: Sage Publications Inc;
Interviewing. Thousand Oaks, CA: Sage Publications Inc; 1996. 1998:21.
18. Campbell DT. Relabeling internal and external validity for 28. Henwood K, Pidgeon, N. Grounded theory. In: Breakwell GM,
applied social scientists. In: Trochim W, ed. Advances in Hammond S, Fife-Schaw C, Smith JA, eds. Research Methods
Quasi-Experimental Design and Analysis. San Francisco: in Psychology. Thousand Oaks, CA: Sage Publications Inc;
Josse-Bass; 1986:67-77. 2006:342–364.
19. Trochim WMK. Research Methods: The Concise Knowledge 29. Stanley M. A grounded theory of the well being of older
Base. Cincinnati, OH: Atomic Dog; 2005. people. In: Finlay L, Ballinger C, eds. Qualitative
20. Daly J, Willis K, Small R, et al. A hierarchy of evidence for Research for Allied Health Professionals: Challenging
accessing qualitative health research. J Clin Epidemiol. Choices. West Sussex, England: John Wiley & Sons;
2007;60:43-49. 2006:63-78.