Qualitative research has recently become more popular in health care areas. This trend becomes apparent as more articles using qualitative research methods are published along with the publication of a new journal (Qualitative Health Research) and a number of new books (e.g., Miles and Huberman, 1994). Historically, qualitative research has never been as popular as quanitative research, and in our view this pattern will not change in the forseeable future. However, qualitative research is unique in terms of its design, data collection and data analysis. For some studies qualitative research is more appropriate than quantitative research. It should be noted that some people believe that doing qualitative research is much easier than doing quantitative research. We believe that this view is not correct. Although qualitative research might be more flexible in terms of research design and possible data collection procedure, an experienced qualitative researcher requires years of training and active participation in the field.
The differences between quantitative and qualitative research are many, ranging from data collection and analysis, to the interpretation of the results. However, the main feature that distinguishes qualitative from quantitative research lies in the nature of the data derived and the analytic process associated with it. Qualitative research often uses data in words (i.e., he is depressed) whereas in quantitative research the data is a number (the systolic blood pressure reads 95 mm Hg). Qualitative research takes a “systemic” approach to understand the interaction of variables in a complex environment. Quantitative research, on the other hand, takes an “analytic” approach to understand a few controlled variables. The former usually conveys a sense of being descriptive, whereas the latter is more analytical.
Qualitative researchers usually select only a small sample of people, or a single case, nested in complex context and studied in-depth. Quantitative research aims primarily at larger numbers of context-stripped cases while seeking statistical significance. A good example of the difference between qualitative and quantitative research in a mailed survey study would be that, in the former, an open-ended question would be used to elicit responses reflecting individual’s concerns, feelings, and experience; in the latter, closed-ended categories, which might be rigid and predictable, are used. The latter are more quantifiable and thus easily interpreted. Qualitative samples tend to be purposive, rather than random.
According to Miles and Huberman, the strength of qualitative data is that it is rich and holistic with strong potential for revealing complexity nested in a real context. Michael, Quinn and Patton summarized the ten common themes of qualitative inquiry:
A type of study that is particularly appropriate for qualitative research is process study and evaluation where the focus of the study is on how something happens rather than on the outcome obtained. This kind of study requires a detailed description of the process, which is usually fluid and dynamic, involving the investigator’s active participation and perception during the process. Also appropriate for a qualitative research are case studies, quality assurance and quality enhancement. Quality assurance (QA), for instance, usually measures the quality of care given to individual clients in order to improve the adequacy and effectiveness of care. The focus on individuals on a case-by- case basis is quite consistent with the use of qualitative methods.
There are generally three steps in data analysis in a qualitative research: data reduction, data display, and conclusion drawing with verification.
Data reduction refers to the process of selecting, and thus simplifying, the data that appears in written field notes or transcriptions. The researcher has to make decisions on how to code the categories, group, and organize them so that the conclusions can be reasonably drawn and verified. Data display refers to ways of displaying the data, which include matrices, graphs, and charts illustrating the patterns and findings from the data. Conclusion drawing and verification refer to a process of developing an initial thought about patterns and explanations from the findings, verifying them constantly by checking the data, and forming a new matrix. It is through such process that the validity of the data is established and the meanings of findings emerge. These three stages of data analysis -- data reduction, data display, conclusion drawing and verification -- form an interactive, cyclical process. As Miles and Huberman illustrate, the coding of data (data reduction) leads to new ideas on what should go into a matrix (data display). As the matrix fills up, preliminary conclusions are drawn, but they lead to the decision, for example, to add another column to the matrix to test the conclusion. In a way, qualitative data analysis is a continuous, iterative enterprise.
For those interested in doing qualitative research, there are two useful books listed under References: Miles and Huberman’s Qualitative Data Analysis : An Expanded Sourcebook and Michael, Quinn and Patton’s Qualitative Evaluation and Research Methods. You may also look into articles published in Qualitative Health Research, a journal obtainable through interlibrary loan.