What is Five-Level QDA all about?

By Nicholas Woolf on May 18, 2016 at 07:30 PM in Five-Level QDA issues & ideas

5LQDA Twitter

by Nick Woolf, 16th May 2016

You may be wondering what Five-Level QDA is all about. Perhaps you’re thinking it’s a new qualitative data analysis (QDA) approach, like discourse analysis, grounded theory or thematic analysis. It’s not. Here in a nutshell is what it’s all about.


Those of us who do QDA, whatever approach we use, all have to choose how to go about it – how to get the thoughts we have when we read data out of our heads and on to … what? Paper, a whiteboard, a set of index cards? Not likely anymore. Almost everyone uses a computer in some way, which is just another kind of tool, like highlighter pens, or scissors to cut up interview transcripts. But … each of those tools automatically elicits an unconscious mindset for what the tool is going to do for you, based on what the tool has done for you before, probably from other contexts. Colored markers segregate things from each other by color in a very clear-cut way, with no sense of overlap between the colored groups of items. Using colored markers to distinguish segments of text on printed-out transcripts therefore may unconsciously nudge you into making clear-cut distinctions between the sets of colored text segments. Nothing wrong with that – if that is your intended analytic strategy.


But if this unconscious nudge from the tool is not the strategy – perhaps your methodology calls for less clear-cut kinds of categories – then there is a conflict – a conflict between the strategies – what you need to do – and the tactics – how you planned to do it. Something similar happens when you open a software program, although the unconscious mindset elicited is very different. Our experience is that with many programs the unconscious mindset is some variant of “the software has been programmed to know what to do, so if in doubt, I’ll just rely on what it seems to suggest I do”. A good example is tax preparation software. If you don’t really understand what it is doing you usually just assume it knows best without double-checking its convoluted calculations. This mindset, however subtle, is not ideal in qualitative research. We don’t want cut-and-dried software operations designed by someone else to guide or even just subtly influence the choices we make in our highly idiosyncratic, unique, and emergent data analyses.


Five-Level QDA is intended to overcome these tendencies by very consciously distinguishing analytic strategies from software tactics, and providing a specific method for moving from strategies to tactics, never the other way around. This means identifying individual small-scale analytic tasks completely independently of the software’s features, and using a “translation” process to match the task not to the big-picture features of the program, but to the more low-level “components” of the software. So it’s a method of harnessing software tools in the service of analytic tasks – whatever methodology you are using. It is not a new or different analytic approach but a way of unpacking the process that experts already use when they harness the full power of these terrific programs, but without having the software’s features drive the data analysis process. That, in a nutshell, it’s what it’s about.

Further reading

Silver, C. and N. H. Woolf (2015). "From guided instruction to facilitation of learning: The development of Five-level QDA as a CAQDAS pedagogy that explicates the practices of expert users." International Journal of Social Research Methodology 18 (5): 527-543.

Jun 03, 2016 Arrow1 Down Reply

merci! :)


The advanced NVivo workshop was so useful. I now have the skills to enable me to use NVivo much more powerfully. Christina is a brilliant trainer and she adapted the course to suit the requirements of those that attended. Christina also gave everyone individual help with their projects. I could not recommend Christina's course highly enough, or emphasise more the value of this course for researchers working with qualitative data.
Caroline Andow, Postgraduate Researcher
University of Southampton