Handling Qualitative Data: the challenge for teachers and researchers
This is a workshop for teachers and, indirectly, for those they teach. It shouldn’t be necessary! I propose to start by discussing why I think it is, and then work through a series of practical topics in discussion, so we can each share experiences and teaching techniques for addressing them. There will be a handout with some of the methods that have worked for me – and a lot of discussion about those that work for others.
Here’s the context. My assessment of the research world I work in is that the great majority of researchers have huge problems with each stage of a project.
Here are ten assertions for discussion!
- The majority of those doing qualitative research have no training in qualitative research.
- The majority of those who do have training learned of the philosophy of method and/or ways of making data – but not what to do with data if they ever had any.
- Research design is rarely taught and there is a widespread sense that qualitative projects just happen. Researchers need to know:
- Why design is essential and the perils of design-free research.
- What’s necessary for setting up a project before those indepth interviews begin. (And what other ways of making data might be better!)
- The necessity for methodological fit – between the question, the method being used, the data being created, the ways of handling data and the outcome desired and designed.
- There is little in the literature about qualitative data records, and researchers need
- Help with how to make data records that are useful representations of what’s being studied – and awareness of the complexity of analyzing such representations.
- Skills for managing the alarming complexity and volume of qualitative data, and ways of setting a project up so they are ready to manage the data.
- A clear understanding of what’s to be done with the data records, so they can be prepared in such a way that they can be adequately accessed..
- Claiming validity for qualitative studies is an increasing problem. Shifts in methodological writing have taken qualitative research further from comfort with the concepts of reliability and validity. And few researchers seem able to design and set up a project in such a way that they would be confident in making such claims.
- In the absence of teaching of data handling as a challenge, researchers “play it by ear”. Which usually means just coding descriptively. The inevitable result will be a study that is descriptive, not analytical. We are in an age of massive proliferation of descriptive qualitative research. Such studies are not bad in themselves, but they fail in any setting where there are goals of providing new understandings. And this trend poses a huge problem for the legitimacy of qualitative research.
- We need to teach how to get up off data and towards new understanding.
- Courses must assist the student in managing ideas and working with abstractions.
- The bogey of theory (and the related one of misunderstood grounded theory) is probably the worst obstacle to confident research for the most students. We need to deal with it instead of making it more scary. Novice researchers need to understand:
- The levels of theory and ways that researchers construct theory
- The sense in which a theory can be grounded – and the meanings of “grounded theory” – now increasingly interpreted as merely meaning you created some understanding of what you had found.
- In my experience, depressingly few of those who start a qualitative project have any strong sense of outcome – of what they might be trying for, or what would be good enough. This is an ethical problem as well as a practical one.
- Software for qualitative research has transformed the method – but the literature barely recognizes this.
- Most researchers will use software but few are trained in the techniques it permits – and these are largely undocumented, still in oral tradition
- Fear of software and lack of teaching means few users are aware of the traps it poses for the untrained, or ways of avoiding those.
- In the absence of a methodological literature that addresses what can be done with software, the vast majority of users of software use only a minimal range of tools and often use those dysfunctionally. Particularly, we need to explain how to use tools for seeing patterns and searching data – and why these are different tasks.
- Reporting of qualitative projects is widely understood to have no standards. Projects that don’t fit into the mold of science journals appear irresponsible. Achieving the right balance of summary and illustration is rarely taught and few researchers appear confident about the outcomes that are acceptable, the use of quoted material or the standards of adequacy of claims. Written reports and oral presentations thus reinforce the low image of qualitative researching
Lyn Richards, March 26, 2005.