Missing Data in Qualitative Research

I’m currently working with Miroslav Klivansky and Josh Rai on Blue Oxen Associates‘ next research report — an extensive case study of the Blue Oxen Collaboration Collaboratory, to be released next month. The study is based on analysis of the community’s archives correlated with the results of a detailed survey of the community’s participants. The goal of the study is to discuss best practices within this community and to propose a framework for examining communities and collaboration. Internally, this is an opportunity to both improve the collaboratory itself and also refine our research methodology.    (AS)

We spent a significant amount of time developing the survey for the study, which was an amazingly difficult process. We had two goals in designing the survey. First, we wanted to gather information about participant behavior that we couldn’t gather from the data itself. For example, we had know way of knowing how much time each participant spent following the community’s discussion. Second, we were trying to determine whether or not the community had QWAN (Quality Without A Name). The problem with this question, of course, is that you can’t just ask it on the survey and expect to get meaningful responses.    (AT)

While struggling with these problems, Miroslav drew our attention to an article by Supriya Singh and Lyn Richards in a recent issue of Qualitative Research Journal — “Missing data: Finding ‘central’ themes in qualitative research” (v3, n1, pp5-17). The article was therapeutic in that it not only empathized with the challenges we were facing, it identified them as standard steps in the research process. Additionally, the article served as a testament to the NUD*IST qualitative analysis tool (the predecessor to NVivo).    (AU)

Singh and Richards write    (AV)

It is rare to find research accounts that do not make the emergence of a theory appear a smooth, even inevitable process. Our own experiences, and those of our students, have never fitted such smooth images, and in discussions we have often found that others are helped by our accounts of the puzzles and anxieties, and the hard detective work, which we have experienced during the analysis stage when a picture appeared to be emerging, but jigsaw pieces were evidently missing. (6)    (AW)

They then explain that the initial research question will inevitable evolve, and hence, there will always be missing data. They also add that survey questions will not always garner the desired information, and hence, the research process must be iterative in order to fill in the blanks. The authors go on to describe the research process for two studies with which they were involved, and explain how they reacted when they discovered missing data.    (AX)

Both authors used the NUD*IST tool extensively, and apparently, the results of their projects “contributed to the further development of the software” (15). I have not had a chance to experiment with NVivo yet, but I hope to do so soon. It sounds like an intriguing tool.    (AY)