Mixing Your MethodsPosted: Thu 11.03.2011
You may have heard the world is made up of atoms and molecules, but it’s really made up of stories. When you sit with an individual that’s been here, you can give quantitative data a qualitative overlay. – William Turner, 16th Century Scientist and Naturalist
As I mentioned last week, I have been working through a mountain of data that I believe I underestimated as I went in to this project. Having 10 participants for a pilot, I thought that it was going to be too small a sample size, or that it would not yield as much data as I needed. But humans are complex beings. Everyone has a story that at least spans the years they have lived, if not working in the complex histories of families and societies. So, from this one experience, I have many individuals with histories and perspectives that have yielded a ton of great work. So now, the question stands, is how to make sense of the rich qualitative data I’ve collected along with the quantitative that comes from pre-/post-surveys. I realized quickly that I was going to need to look to bolster my understanding of what unique challenges and opportunities mixed methods research poses.
Working on a study that incorporates both qualitative and quantitative methods is nothing new, but it is a science unto itself. After reading through some titles on mixed methods design, my advisor recommended looking in to the work of Tashakkori & Teddlie (2003). Delving into the text, I am struck by how many assumptions I made about mixed methods design. For instance, I am now aware of the fact that I never had a proper definition of what mixed methods design is, which seems obvious, but it also is fraught with assumptions. As the authors cited their previous work (1998), Mixed methods studies are those that combine the qualitative and quantitative approaches into the research methodology of a single study or multiphased study.” This may seem logical, but it is something that I never thought about the need to explicitly state. As my Myers-Briggs test would confirm, I am more of an abstract thinker and that sometimes translates into taking details for granted and moving forward with grander theories without the base work. So, the first benefit of this book is the handbook format that allows me to fill in the gaps of the concrete definitions and processes that I have been floating along on my assumptions. What is clear is that research is an ongoing quality improvement process and one that I strive to refine as I move forward.
Along with the education on terms and frameworks, the handbook has also given me a look into the sophistication of data analysis available for cross-referencing qualitative and quantitative data. The authors assert that when linking empirical with theoretical, it is not enough to employ induction or deduction. Instead, the researcher must incorporate different kinds of triangulation techniques to see if the data converges, complements, or diverges/contradicts. Denzin (1978) calls this process “between-method triangulation”: seeing if analyzing qualitative and quantitative separately reports a different story than when analyzed together. What strikes me about this is the sophistication of analysis methods and the scientific nature of a process that is looking at very social, human data. Such an approach ensures a more complete picture emerging to the researcher and a better understanding of how factors coexist and co-mingle to affect behavior and phenomena.
Overall, I have to say that the book is a clear road map through the complicated, diverse nature of mixed methods research. I did not think that I could find a book I loved more than Patton’s (2002) Qualitative Research & Evaluation Methods, but I am finding this a contender. On a side note, my next step in the process is to see what benefits entering my data into software might yield. I have just signed up for a trial of the software Dedoose and am also trying out NVivo. Depending on how comfortable I get with either or both, a review/comparison will follow.