Many important medical education research questions cry out for a qualitative research approach: How do teacher characteristics affect learning? Why do learners choose particular specialties? How is professionalism influenced by experiences, mentors, or the curriculum? The medical paradigm, the “hard” science most often taught in medical schools, usually employs quantitative approaches.1 As a result, clinicians may be less familiar with qualitative research or its applicability to medical education questions. For these why types of questions, qualitative or mixed qualitative and quantitative approaches may be more appropriate and helpful.2 Thus, we wish to encourage submissions to the Journal of Graduate Medical Education that are for qualitative purposes or use qualitative methods.
This editorial is the first in a series of two, and it will provide an introduction to qualitative approaches and compare features of quantitative and qualitative research. The second editorial will review in more detail the approaches for selecting participants, analyzing data, and ensuring rigor and study quality in qualitative research. The aims of the editorials are to enhance readers' understanding of articles using this approach and to encourage more researchers to explore qualitative approaches.
Theory and Methodology
Good research follows from a reasonable starting point, a theoretical concept or perspective. Quantitative research uses a positivist perspective in which evidence is objectively and systematically obtained to prove a causal model or hypothesis; what works is the focus.3 Alternatively, qualitative approaches focus on how and why something works, to build understanding.3 In the positivist model, study objects (eg, learners) are independent of the researchers, and knowledge or facts are determined through direct observations. Also, the context in which the observations occur is controlled or assumed to be stable. In contrast, in a qualitative paradigm researchers might interact with the study objects (learners) to collect observations, which are highly context specific.3
Qualitative research has often been differentiated from quantitative as hypothesis generating rather than hypothesis testing.4 Qualitative research methods “explore, describe, or generate theory, especially for uncertain and ‘immature’ concepts; sensitive and socially dependent concepts; and complex human intentions and motivations.”4 In education, qualitative research strives to understand how learning occurs through close study of small numbers of learners and a focus on the individual. It attempts to explain a phenomenon or relationship. Typically, results from qualitative research have been assumed to apply only to the small groups studied, such that generalizability of the results to other populations is not expected. For this reason, qualitative research is considered to be hypothesis generating, although some experts dispute this limitation.5 table 1 presents a comparison of qualitative and quantitative approaches.
When Qualitative Studies Make Sense
Qualitative studies are helpful to understand why and how; quantitative studies focus on cause and effect, how much, and numeric correlations. Qualitative approaches are used when the potential answer to a question requires an explanation, not a straightforward yes/no. Generally, qualitative research is concerned with cases rather than variables, and understanding differences rather than calculating the mean of responses.4 In-depth interviews, focus groups, case studies, and open-ended questions are often employed to find these answers. A qualitative study is concerned with the point of view of the individual under study.6
For example, the changes in duty hours for residents in 2003 have generated many quantitative research articles, which have counted and correlated the changes in numbers of procedures, patient safety parameters, resident test results, and resident sleep hours. However, to determine why residents still sleep about the same number of hours since 2003, one could start from a qualitative framework in order to understand residents' decisions regarding sleep. Similarly, to understand how residents perceive the influence of resident work hour restrictions on aspects of professionalism, a qualitative study would start with the learners rather than by measuring and correlating scores on professionalism assessments. Because learning takes place in social environments characterized by complex interactions, the quantitative “cause and effect” model is often too simplistic.7
A variety of ways to collect information are available to researchers, such as observation, field notes, reflexive journals, interviews, focus groups, and analysis of documents and materials; table 2 provides examples of these methods. Interviews and focus groups are usually audiorecorded and transcribed for analysis, whereas observations are recorded in field notes by the observer.
After data collection, accepted methods are employed to interpret the data. Researchers review the observations and report their impressions in a structured format, with subsequent analysis also standardized. table 3 provides one example of an analysis plan. Strategies to ensure rigor in data collection and trustworthiness of the data and data analysis will be discussed in the second editorial in the series.
In contrast to quantitative methods, subjective responses are critical findings, both in participant responses and observer reactions. The unique or outlier response has value in contributing to understanding the experience of others, and thus individual responses are not lost in the aggregation of findings or in the development of research group consensus.2,4 Qualitative methods acknowledge the “myth of objectivity” between researcher and subjects of study.7 In fact, the researcher is unlikely to be a purely detached observer.
As qualitative researchers usually attempt to study subjects and interactions in their “natural settings,” ethical issues frequently arise. Because of the sensitive nature of some discussions as well as the relationship between researchers and participants, informed consent is often required. The very reason for doing qualitative research—to discover why and how, particularly for thorny topics—can lead to potential exposure of sensitive opinions, feelings, and personal information. Thus, consideration of how to protect participants from harm is essential from the very onset of the study.
Qualitative researchers need to show that their findings are credible. As with quantitative approaches, a strong research project starts with a basic review of existing knowledge: a solid literature search. However, in contrast to quantitative approaches, most qualitative paradigms do not look to find a single “truth,” but rather multiple views of a context-specific “reality.” The concepts of validity and reliability originally evolved from the quantitative tradition, and therefore their accepted definitions are considered inadequate for qualitative research. Instead, concepts of precision, credibility, and transferability are key aspects of evaluating a qualitative study.9
Although some experts find that reliability has little relevance to qualitative studies, others propose the term “dependability” as the analogous metric for this type of research. Dependability is gained though consistency of data, which is evaluated through transparent research steps and research findings.9,10 Trustworthiness and rigor are terms used to establish credible findings. One technique often used to enhance trustworthiness and rigor is triangulation, in which multiple data sources (eg, observation, interviews, and recordings), multiple analytic methods, or multiple researchers are used to study the question.9 The overall goal is to minimize and understand potential bias while ensuring the researcher's “truthfulness” of interpretation.9
A potentially helpful appraisal checklist for qualitative studies, developed by Coté and Turgeon,11 is found in table 4. This appraisal checklist has not been examined systematically. table 5 includes a list of terms commonly used in qualitative research. Approaches to ensure rigor and trustworthiness in qualitative research will be addressed in greater detail in Part 2.
Both quantitative and qualitative approaches have strengths and weaknesses; medical education research will benefit from each type of inquiry. The best approach will depend on the kind of question asked, and the best methods will be those most appropriate to the question.4 To learn more about this topic, the references below are a useful start, as is talking to colleagues engaged in qualitative research at your institution or in your specialty.
Gail M. Sullivan, MD, MPH, is Editor-in-Chief, Journal of Graduate Medical Education; and Joan Sargeant, PhD, is Professor in the Division of Medical Education, Dalhousie University, Halifax, Nova Scotia, Canada.