How Will We Know "Good" Qualitative Research When We See It? Beginning the Dialogue in Health Services Research

Health Services Research, Dec, 1999 by Kelly J. Devers

The desire to reduce barriers to combining qualitative and quantitative methods requires further comment given the interest in mixed methods, and their prevalence, in health services and policy research. I discuss this subject briefly here, before moving to a discussion of two forces that have stimulated other disciplines to reconsider the use of broad, positivist criteria in evaluating qualitative research.

In their now classic work entitled "Beyond Qualitative Versus Quantitative Methods," Reichardt and Cook (1978) argue that the advantages of combining qualitative and quantitative methods should compel disciplines (in their article, the field of evaluation research) to move beyond the traditional qualitative versus quantitative debate to an acceptance of both methods. The advantages they describe include the multiple purposes for which research is conducted, the complementary aspects of quantitative and qualitative methods, and triangulation (in this case, using more than one method to verify and validate results). Consequently, they conclude, researchers should use whatever methods are best suited to their research needs, regardless of their traditional methodological affiliations. Although the authors acknowledge that obstacles to combining qualitative and quantitative methods exist (i.e., time, money, lack of multi-methods training, and funding "fads"), they argue that in many circumstances the benefits fa r outweigh the costs.

Moreover, Reichardt and Cook (1978) contend that it is the confusion about the "real, but imperfect" link between paradigms and methods that increases the conflict between qualitative and quantitative researchers and that prevents them from combining these methods where useful. Methods and paradigms are logically separable. Although it is true that agreement between paradigm and method is frequent, other combinations are possible. [10]

Although the Reichardt and Cook (1978) article helped move the basic and applied social sciences from a "quantitative versus qualitative" paradigm debate to a "quantitative and qualitative" methods detente, it did not explicitly address two questions that have triggered and intensified arguments in the social sciences today. The first question is, What paradigm is adopted when qualitative and quantitative methods are combined? The authors do not explicitly state this, but the paradigm most frequently adopted when methods are combined is that associated with quantitative research. This includes the adoption of broad criteria that are traditionally used to evaluate quantitative research, as described earlier (see Figure 1). The only hint the authors provide that a shared paradigm might be problematic is in their brief discussion about how its absence among multi-disciplinary team members can impede the successful completion of a research project. (See Ianni and Orr 1978 and

Daly and McDonald 1992 on the challenges of working in multi-disciplinary teams; four recent articles on combining qualitative and quantitative research and methods in health services research are Barbour 1999; Morgan 1998; Stange, Miller, Crabtree, et al. 1994; and Ward 1993).