WHAT ARE MIXED METHODS?
Mixed methods research is characterized by the combination of quantitative and qualitative methods to address the depth (e.g., individual perspective on why an intervention does or does not work) and breadth (e.g., mediation analysis) of research questions ( Johnson, Onwuegbuzie, & Turner, 2007). Mixed methods research involves the integration of quantitative and qualitative approaches to study designs, data collections, and data analyses. Some “qualitative” approaches involve numeric methods (e.g., cultural consensus analysis; Romney, Batchelder, & Weller, 1987), while some “quantitative” approaches based on statistics involve qualitative judgments (e.g., the number of classes in a latent class model; van Smeden, Naaktgeboren, Reitsma, Moons, & de Groot, 2014). We should state at the outset that the terms “quantitative” and “qualitative” are shorthand for broad approaches often characterized as “numeric” (statistical, objective) or “text” (interpretive, subjective). Depending on the type of mixed methods study, quantitative or qualitative methods may have a more primary role, with an emphasis on the respective theoretical assumptions, study designs, and analytical approaches.
Quantitative and qualitative frameworks have developed from distinct disciplinary worldviews (DePoy & Gitlin, 2015). Quantitative methods are derived from positivist assumptions that are ideal for establishing cause-and-effect relationships (determinism), identifying key variables to describe a phenomenon (reductionism), measuring a construct (measurement), or testing a hypothesis (deductive logic)— an etic or “culturally neutral” perspective. Qualitative approaches are framed from constructivist or realistic worldviews that seek insight and interpretation of the context in which interventions take place at individual, social, and organizational levels—an emic or “culturally unique” perspective (Robins et al., 2008). Often quantitative approaches (exemplified by the fields of biostatistics and epidemiology) seek to generalize replicable results from the sample to the population (e.g., using a standardized depression questionnaire to estimate the prevalence of depression in the population from results in a sample). In contrast, qualitative approaches (exemplified by the fields of anthropology and sociology) seek to understand the scope of a domain (e.g., understanding the experience and concept of depression from the point of view of an individual from a certain culture).
Mixed methods maximize the strengths and counterbalance the weaknesses of quantitative and qualitative approaches in a single study or program of research. Purposeful and planned integration of quantitative and qualitative approaches is a key feature of mixed methods. Thinking carefully about the purposes and stages of intervention design allows investigators to build strong ties between the research question, methods, and theory. Driven by the needs and goals of the research, investigators should make specific and planned efforts to integrate quantitative and qualitative methods at all stages of the pipeline—study design, data collection, analysis, and interpretation. In the rest of this chapter, we describe how mixed methods can be used across the intervention pipeline to the design and development and evaluation of interventions to implementation in community and clinical settings.