Combining Qualitative and Quantitative
Research Methods: Considering the Possibilities for Enhancing the Study of
This paper discusses some of the underlying reasons why
health researchers have historically had difficulty working
collaboratively across qualitative and quantitative research paradigms and
argues why it is imperative that researchers move beyond traditional
adherence to particular methods of inquiry. Chronic illnesses are prime
examples of conditions that by their very nature need to be studied from a
combination of perspectives, using both qualitative and quantitative
methods. We suggest that the success of health research on managing these
conditions lies in the shared application of both qualitative and
quantitative research perspectives, methods and tools. In addition, we
argue that effective research into long-term chronic illnesses requires
not only combined research efforts but also longitudinal programs of
study, so that the experience of managing chronic conditions can be
captured over time.
Individuals experiencing the symptoms of chronic illnesses often struggle to accept and live with a given condition long before the illness has been given a name and in advance of seeking treatment. Chronic diseases are usually diagnosed and named by physicians; this often leads to long-term treatment and monitoring of activities by a range of clinicians and other caregivers involved in the management of such conditions. In order to properly understand and cope well with chronic, long-term and often increasingly debilitating illnesses, a broad range of knowledge and understanding must be brought together over time to support both the quality and quantity of remaining life for the individual living with the disease.
In particular, research that helps extend our understanding of how best to manage such chronic diseases and corresponding illness experiences requires a broad range of perspectives and skills. Holman1 delineates a clear need for incorporating qualitative inquiry into medical research by highlighting the case of chronic diseases. He concludes that “good medical research recognizes the complementarity and interpretation of quantitative and qualitative methods of inquiry.”
Unfortunately, the ability to combine research expertise across traditional methodological boundaries is often thwarted. Qualitative and quantitative researchers often operate with a different set of assumptions about the world and ways of learning about it. These assumptions may be seen as mutually and inevitably irreconcilable. Researchers are often taught to master only one type of method and, so, become comfortable with their expertise in handling either quantitative or qualitative analysis, but not both. The result is that the two major approaches (qualitative and quantitative) are seldom combined and their respective strengths are ignored by adherents of each approach.2
A review of the chronic disease literature confirms the general trend to conduct either quantitative or qualitative studies, identifying a relative paucity of combined method designs. A review of MEDLINE citations for the period 1993 to September 1997 indicates that 305 quantitative studies of chronic diseases were published, while only 112 qualitative studies were published (providing numerical evidence, at least, of a continuing preference for quantitative analysis). In all, 47 papers citing both quantitative and qualitative techniques were referenced. Many of these studies, however, simply used qualitative diagnostic measurements within essentially quantitative, quasi-experimental designs or else included reviews of both qualitative and quantitative literature relevant to the chronic condition of interest. Only 13 could truly be categorized as combined method studies.
Whether this is evidence of an active debate or simply the existence of two separate tracks of research efforts, the use of one method or the other clearly remains the predominant approach to the study of chronic diseases. This trend mirrors the situation with respect to health research in general. The inherent danger in this strict separation of research perspectives is the likely production of incomplete results regarding the health problem being studied.
In an attempt to encourage the conduct of combined method analyses, this paper does the following.
Quantitative research is defined as “the numerical representation and manipulation of observations for the purpose of describing and explaining the phenomena that those observations reflect,” and qualitative research is described as “the non-numerical examination and interpretation of observations, for the purpose of discovering underlying meanings and patterns of relationships.”3 Reviewing these definitions of what is meant by quantitative versus qualitative research helps identify the reasons for the primarily separate use of each method and the continuing debate among researchers concerning the relative value of each approach. The arguments can be complicated and often are philosophical; however, they essentially make the following kinds of distinctions.
While it may be somewhat naive to delineate the differences between qualitative and quantitative research so definitively, it is helpful to begin to understand the nature of the debate by understanding commonly held divisions and basic definitions. The basic constructs for viewing what “scientific” research is too often divide researchers in the health field, where the clinical trial remains the gold standard against which all other research is bench-marked.5 Unfortunately, these definitions tend to establish two separate and contrary schools of research, emphasizing the arguments commonly engaged in to justify the use of one or the other technique, rather than simply stating the varying positions and perspectives contained within qualitative and quantitative research paradigms.
The Usual Distinctions
Quantitative and qualitative research methods are most often associated with deductive and inductive approaches, respectively. Deductive research begins with known theory and tests it, usually by attempting to provide evidence for or against a pre-specified hypothesis. Inductive research begins by making observations, usually in order to develop a new hypothesis or contribute to new theory. Quantitative research is usually linked to the notion of science as objective truth or fact, whereas qualitative research is more often identified with the view that science is lived experience and therefore subjectively determined. Quantitative research usually begins with pre-specified objectives focused on testing preconceived outcomes. Qualitative research usually begins with open-ended observation and analysis, most often looking for patterns and processes that explain “how and why” questions.
When applying quantitative methods, numerical estimation and statistical inference from a generalizable sample are often used in relation to a larger “true” population of interest. In qualitative research, narrative description and constant comparison are usually used in order to understand the specific populations or situations being studied. As a result, quantitative research is most often seen as a method trying to demonstrate causal relationships under standardized (controlled) conditions. Conversely, qualitative research is usually seen as a method seeking better understanding of some particular, natural (uncontrolled) phenomenon. A summary of the kinds of distinctions often made concerning the use and value of both methods is provided in Table 1.
The nature of the general theoretical debate, then, is characterized by fundamentally different understandings or beliefs about scientific research, in particular, and the world, in general. Adherence to different and separate paradigms can trap researchers into believing that there is only one true “scientific” way to conduct research.6 Exceptions to the general rules or tenets associated with these research approaches suggest that many of the clashes between researchers' perspectives are more a question of belief systems and mutual lack of understanding than of methods. Nonetheless, the arguments continue to focus on methodological aspects.
Clarifying Differences and Acknowledging Similarities
The dichotomy of quantitative, deductive analysis under standardized, objective conditions versus qualitative, inductive inquiry aimed at understanding phenomena in uncontrolled, natural contexts remains a barrier between researchers from different analytical disciplines,7 particularly those studying the etiology and consequences of disease.8 We believe these distinctions are particularly unhelpful when the target of research is the study of chronic problems. Chronic diseases, by their very nature, require the complementary use of qualitative and quantitative research methods in order to quantify the effectiveness of treatments and qualify the illness experience as it progresses over time.1
Instead of either ignoring or defending a particular research paradigm, it is possible and more instructive to see qualitative and quantitative methods as part of a continuum of research techniques, all of which are appropriate depending on the research objective. For example, Shaffir and Stebbins have modeled this continuum in a way that challenges the notion that qualitative approaches are solely exploratory and inductive, while quantitative methods are only explanatory and deductive.9 Guba and Lincoln,10 offer this comment.
Both qualitative and quantitative methods may be used appropriately with any research paradigm. Questions of method are secondary to questions of paradigm, which we define as the basic belief system or world view that guides the investigator.
Careful review of the full spectrum of both major research paradigms will confirm that both methods can be used in less “usual” ways, i.e. it is possible to quantitatively describe observable events in the real world and to collect qualitative evidence within pre-specified, experimental situations. Alongside recognizing that both methods can be used in these “unusual” ways, it is also important to remember that both methods contain many different approaches. For example, grounded theory and case study are different approaches than those of ethnography or phenomenology, and yet all four approaches are essentially qualitative. The same sort of distinctions apply to quantitative approaches: all clinical trials are not identical in design and therefore use differing techniques for measuring results, and there are many different forms of experimental, quasi-experimental and pre-experimental designs, using equally varied quantitative analyses. The point here is not to understand the specific differences of these techniques, but to highlight the existence of a range of options under both the qualitative and quantitative “umbrellas.”11–13
There is a need to recognize that both methodological schools have an equally respectable place in health and health care research; quantitative and qualitative techniques can and should co-exist as potential tools of the research trade. Instead of worrying about justifying the less highly regarded method (which appears to shift over time and across disciplines in any case), efforts should focus on understanding why and when to use one or the other method, or both.
The concept of capturing the inter-relatedness, rather than the differences, of qualitative and quantitative methods is beginning to receive greater attention in health research. For example, Daly et al.14 indicate that effective health care research cannot be reduced to a matter of using a “strong” [usually thought to be quantitative rather than qualitative] method; rather, many important health issues must be tackled using a range of methods.
In short, our belief is that the broad range of questions that arise from complex health care problems can only adequately be addressed by an equally broad range of research study designs. Those who undertake evaluation and research in health care must therefore cultivate methodological flexibility.
Any study design or combination of research methods selected for use should be responsive to the particular research problem or question. For example, exploring the implementation of new health care delivery arrangements requires a mainly qualitative approach, since it is not possible to divorce the processes of change under study from the social contexts in which they occur. On the other hand, a study designed to identify the glucose tolerance of people with diabetes versus people without the disease requires essentially quantitative analysis of differences under carefully controlled conditions. Approaching the question of which method is best in this way avoids the debate entirely and, arguably, better conforms to the rules of “good” science.15
Collaborating Across Traditional Boundaries
Within the discussion of what constitutes good science, there is a slow but important movement toward more collaborative use of both types of research methods in the field of health research, particularly in relation to the study of chronic diseases. Positive suggestions for combining quantitative and qualitative approaches are emerging from some health-oriented disciplines.
Sociology and nursing are fields that struggle with the divide often separating researchers who prefer one or the other technique; however, some researchers in both fields are promoting greater harmony. For example, sociologists Strauss and Corbin16 stated: “To systematize and solidify connections we use a combination of inductive and deductive thinking, in which we constantly move between questions, generating hypotheses and making comparisons.” Commenting on the schism in nursing research, Corner17 suggested that “the use of different research methods within a single study can provide a richer and deeper understanding of the area under investigation than would otherwise be possible.”
Medical research is also beginning to include qualitative approaches more often. For example, a team of physicians and other researchers18 used a qualitative inquiry nested within a larger clinical survey in order to better elicit patient expectations for medical care.
Clearly, some developments across a continuum of health care research follow the logic of identifying what kinds of research skills and corresponding methods are responsive to the problems that require addressing within the health care field today. The capacity to better understand and improve the ongoing support for individuals living with chronic diseases surely encompasses a number of such health and health care challenges.
Examples from Chronic Disease Research
Spanning Paradigms and Combining Methods
Combined method research approaches are particularly suited to the study of chronic disease and long-term illness.1,19–21 It is especially important to recognize the benefits of widening research efforts in relation to chronic diseases. As Holman1 comments," Conventional biomedical research has not provided decisive information about the origins or management of the most prevalent contemporary medical problems, namely, chronic illnesses." Finding ways to move beyond the limitations of traditional research boundaries can help expand understanding of some of the most long-term and widespread health problems facing populations today. We believe that it is through combined, sustained and complementary use of qualitative and quantitative research methods that advances in our knowledge of chronic diseases can best be attained.
Once the philosophical differences (different world views) and practical barriers (lack of knowledge or expertise) to using a combined approach are recognized, they can be managed constructively, and a number of potential ways to usefully combine qualitative and quantitative techniques will emerge. Various combinations have been described by several health researchers;22–25 however, use of these potential multi-method options is still relatively new in the context of chronic disease research.
Some examples of how combined study approaches can help us better understand and treat individuals living with chronic diseases make the generic uses outlined in Table 2 more accessible and relevant. Six possible uses of combined method approaches are described below to help promote their use within the context of future chronic disease research.
Use #1: To Develop Measures
The most generally accepted use of combined methods is to begin with a qualitative exploration of some little-studied problem so that measurement instruments can be developed for later quantitative research. For example, Bauman and Adair's study26 of social support among inner-city mothers of children with chronic illnesses used qualitative interviewing to inform the construction of a questionnaire. An exploratory study of how chronic diabetes affects quality of life and/or treatment choices among cultural groups not yet well researched (in order to subsequently develop a survey tool) provides another example well-suited to this design type.
Use #2: To Identify Relevant Phenomena
Persons coping with chronic conditions often rely on the effectiveness of medications to relieve or at least minimize painful, debilitating symptoms. Such is the case for individuals with heart disease who control pain from angina with prescribed medication. Quantitative research has identified drugs that are “effective” for this pain control. Such study alone, however, does not always sufficiently describe all of the side effects that may accompany this pain relief, nor can it encompass the meaning for individuals suffering from these side effects. The addition of qualitative study can often more fully identify and explain side effects or problems of compliance with drug regimes experienced by people living with chronic heart disease. Clearly, both kinds of evidence are critical if research is to capture the full experience of long-term chronic heart disease.
Additional examples from the literature include a study by Bashir et al.,27 looking at the relevance of qualitative advice to chronic benzodiazepine users, and work by Borges and Waitzkin.28 The latter researchers conducted a review of both quantitative and qualitative patient-doctor communication techniques in order to develop an interpretive method for their study of women with chronic social and emotional problems.
Use #3: To Interpret/Explain Quantitative Data
The quantification of rates of chronic disease can often leave researchers with unanswered questions about why rates are different over time or by geographic region. So, for example, when quantitative evidence points to a seemingly unexplained high prevalence of asthma in Alberta compared to other parts of Canada, qualitative analysis of the reasons for this is warranted. Qualitative techniques will tell the story behind the comparative quantification. Wainwright's study29 of chronic liver disease employs this approach, describing how qualitative research can lead to additional quantitative and qualitative appraisal of psychological adjustment to end-stage chronic liver disease.
Use #4: To Interpret/Explain Qualitative Data
The exploration of qualitative aspects of living with many chronic conditions can lead to a deep understanding of how certain individuals experience living with their illnesses. Sometimes, these descriptive data on how people live with a particular chronic condition appear inconsistent according to gender or age differences. In the context of a qualitative study, both sample size and method are inadequate to test the validity of any apparent distinctions. Only quantitative study can test these findings with sufficient and appropriate sampling.
In Finkler and Correa's study30 of patients' perceived recovery and the role of the patient-doctor relationship, statistical analysis revealed that only some components of the relationship significantly influenced treatment outcomes. This prompted review of the qualitative data and further development of a qualitative understanding of the patient-doctor relationship.
Use #5: To Gain Equal/Parallel Value from Both Types of Data
The above reason for combining methods is arguably the most prevalent in today's climate of continuing scepticism about use of both qualitative and quantitative approaches in single studies. Examples could include combining the results of any studies that separately, but simultaneously, research a chronic disease. So, while one group of researchers is quantifying the efficacy of a particular treatment for breast cancer, another group may be comparing women's experiences of surviving following detection and treatment of breast cancer. In this case, however, a combined, co-ordinated study design has not really been accomplished.
Truly combined method approaches would purposely connect the study objectives and methodologies in the context of a single study or within a planned program of research in order to access a more comprehensive range of information and experience. For example, Martin and Nisa's research31 intentionally combines qualitative and quantitative information to describe common features of common chronic children's illnesses. Rutgaizer and Larina's study32 of pain syndrome in gastroenterological practice, and Murray and Graham's work33 on community health needs also use a truly combined method approach.
Use #6: To Conduct Effective Multistage (Longitudinal) Analysis
The very nature of chronic illness invites longitudinal programs of research.1,34 Long-term illness often entails coping with a number of increasingly disabling stages; arguably, each stage requires understanding in and of itself and in the context of the overall condition. A program of ongoing study, using both qualitative and quantitative research, can provide essential knowledge of the changing nature of the disease and the corresponding experience and needs of persons with the illness.
So, for example, a survey of persons living with HIV/AIDS, aimed at quantifying the difference that new treatments are making in relation to longevity, might be planned in the context of a follow-up study focused on qualitatively assessing changes in quality of life. These study phases could, in turn, be followed by an experimental intervention designed to compare the efficacy of two different treatment regimes, over time and in relation to length and quality of life.
A research program conducted by Bates and Rankin-Hill,35 combining two qualitative and two quantitative projects among chronic pain sufferers, provides a recent example of this longitudinal approach to combined use of methods. The “spiral” approach used by de Vries et al.36 in their development of health education program planning also describes the valuable “interaction” that use of both qualitative and quantitative methods over time can achieve.
Making a Difference through Multi-method Study
There is an intuitive appeal to combining research approaches, but the barriers separating researchers— different world views, different training, simple lack of contact and understanding—all conspire to make collaboration difficult. Perhaps discussion, education and debate of the advantages and potentials for working across research paradigms will find a particularly receptive audience in relation to chronic disease conditions. If research can be improved by linking quantitative and qualitative researchers and using a combined method approach to the study of managing chronic disease, then such collaborations should be encouraged and supported. People living with long-term illnesses surely deserve the best efforts of the research community.
1. Holman HR. Qualitative inquiry in medical research J Clin Epidemiol 1993;46(1):29–36.
2. Polit DF, Hungler BP. Nursing research: principles and methods. 4th ed. New York (NY): JB Lippincot Company, 1991.
3. Babbie E. The practice of social research. 6th ed. Belmont (CA): Wadsworth Publishing Company, 1992.
4. Denzin NK, Lincoln YS, editors. Handbook of qualitative research. Thousand Oaks (CA): Sage Publications, 1994.
5. Popper KR. The logic of scientific discovery. New York (NY): Routledge, 1980.
6. Verhoef MJ, Casebeer AC. Broadening horizons: integrating quantitative and qualitative research. Can J Infect Dis 1997;8(2):665–6.
7. Chalmers AF. What is this thing called science? An assessment of the nature of science and its methods. 2nd ed. Queensland (Australia): University of Queensland Press, 1982.
8. Feinstein AR. An additional basic science for clinical medicine: I. The constraining fundamental paradigms. Ann Intern Med 1983;99:393–7.
9. Shaffir WB, Stebbins RA, editors. Experiencing fieldwork: an insiders view of qualitative research. Newbury Park (CA): Sage Publications, 1991.
10. Guba EG, Lincoln YS. Fourth generation evaluation. Newbury Park (CA): Sage Publications, 1989.
11. Mays N, Pope C, editors. Qualitative research in health care (a series of seven papers initially published in the BMJ). London (UK): BMJ Publishing Group, 1996.
12. Campbell DT, Stanley JC. Experimental and quasi-experimental designs for research. Chicago (IL): Rand McNally, 1963.
13. Creswell JW, Research design: qualitative and quantitative approaches. Thousand Oaks (CA): Sage Publications, 1994.
14. Daly J, McDonald I, Willis E. Researching health care: designs, dilemmas, disciplines. London (UK): Routledge, 1992.
15. Hulley SB, Cummings SR, editors. Designing clinical research. Baltimore (MD): Williams and Wilkins, 1988.
16. Strauss AL, Corbin J. Basics of qualitative research: grounded theory procedures and techniques. Newbury Park (CA): Sage Publications, 1990.
17. Corner J. In search of more complex answers to research questions: qualitative versus quantitative research methods: is there a way forward?. J Adv Nurs 1991;16:718–27.
18. Kravitz RL, Callahan EJ, Paterniti D, Antonius D, Dunham M, Lewis CE. Prevalence and sources of patients' unmet expectations for care. Ann Intern Med 1996;125(9):730–7.
19. Bluebond-Langner M, Perkel D, Goertzel T, Nelson K, McGeary J. Children's knowledge of cancer and its treatment: impact of an oncology camp experience. J Pediatr 1990;116(2):207–13.
20. Jamison RN, Lewis S, Burish TG, et al. Psychological impact of cancer on adolescents: self image, locus of control, perceptions of illness and knowledge of cancer. J Chronic Dis 1986;39:609–17.
21. Hilsden RJ, Verhoef MJ, Scott CM. Complementary therapy use by patients with inflammatory bowel disease. Am J Gastroenterol. In press 1997.
22. Steckler A, Mcleroy KR, Goodman RM, Bird ST, McGormich L. Toward integrating qualitative and quantitative methods: an introduction. Health Educ Q 1992;19:1–8.
23. Feinstein AR. An additional basic science for clinical medicine: III. The challenges of comparison and measurement. Ann Intern Med 1983;99:705–12.
24. Mays N, Pope C. Rigour and qualitative research. BMJ 1995;311:109–12.
25. Afifi AA, Breslow L. The maturing of public health. Annu Rev Public Health 1994;15:223–35.
26. Bauman LJ, Adair EG. The use of ethnographic interviewing to inform questionnaire construction. Health Educ Q 1992;19(1):9–23.
27. Bashir K, King M, Ashworth M. Controlled evaluation of brief intervention by general practitioners to reduce chronic use of benzodiazepines. Br J Gen Pract 1994;44:408–12.
28. Borges S, Waitzkin H. Women's narratives in primary care medical encounters. Women Health 1995;23(1):29–56.
29. Wainwright SP. Transcending chronic liver disease: a qualitative study. J Clin Nurs 1997;6(1):43–5.
30. Finkler K, Correa M. Factors influencing patient recovery in Mexico. Soc Sci Med 1996;42(2):199–207.
31. Martin C, Nisa M. Meeting the needs of children and families in chronic illness and disease: a greater role for the GP? Aust Fam Physician 1996;25(8):1273–81.
32. Rutgaizer IM, Larina NA. Pain syndrome in gastroenterological practice: potentialities of evaluation [Russian: abstract only] Klin Med 1996;74(4):45–8.
33. Murray SA, Graham LJ. Practice based health needs assessment: use of four methods in a small neighbourhood. BMJ 1995;310(6992):1443–8.
34. Huber GP, Van de Ven AH, editors. Longitudinal field research methods: studying process of organizational change. Thousand Oaks (CA): Sage Publications, 1995.
35. Bates MS, Rankin-Hill L. Control, culture and chronic pain. Soc Sci Med 1994;39(5):629–45.
36. de Vries H, Weijts W, Dijkstra M, Kok G. The utilization of qualitative and quantitative data for health education program planning, implementation and evaluation: a spiral approach. Health Educ Q 1992;19(1):101–15.