I recently gave a short presentation to some students about qualitative data analysis, and the discussion that followed inspired me to write some more about this subject here.

There are many approaches to analyze qualitative data, and each researcher normally crafts the most appropriate for her or his project.  Using available transcripts, field notes, photos and video material, a common way to proceed looks like this:

find themes —>  create codes to organize them —> find interrelationships and connections between themes —> combine

A thing to keep in mind, however, is that data collection and analysis often go on simultaneously. The actual process is not quite as simple and linear as the one pictured above. Analysis involves both finding patterns and making patterns. Some analysis work resembles the process of sewing up a quilt — cutting small bits and rearranging them in a certain way to create a new, symmetrical, complex pattern. But before starting our quilt/analysis process it is very necessary to ask ourselves: what is it for?


This is where the conversation with the students got interesting, and made me wonder whether I shouldn’t have started my discussion by addressing exactly that query: what are you analyzing for? Some researchers are developing a theory, or a model. In other cases the outcome need be a solid set of insights for a design project, as it was the case for many of these students. Or the creation of a new policy, the basis for a development project. It could also be that the task at hand is to ‘simply’ identify a problem and dissect it, or forecast trends (although in this last case we are talking more market analysis, which often involves quantitative more than qualitative data).

While collecting data and analyzing them, keep in mind your final outcome (design project, theoretical model, etc) and always remember to craft relevant ways to enrich your data collection process. Experiment a bit, and try to talk to other researchers to bounce off ideas.

My master course taught us — nearly forced us, really, to use a strong word — mostly to develop theories from the analysis of our ethnographic data. No matter how many other theories out there were already tackling our same issues in more articulated and informed ways, we had to set out in the world to develop a theory. I soon realized that I found it a bit more inspiring to analyze data in order to point to solutions to practical problems and/or to contribute to the design of a service/product/initiative. That way connecting the dots becomes creative and a little less ‘logical’. Leaps of imagination and disruption are key here, following the logical path doesn’t necessarily lead to a great solution. Ok, but how do you do that? How do we analyze data to produce insights that point to good solutions?

The first answer is patience, trial and error and a willingness to be more ‘quick and dirty’ than the regular anthropological praxis would allow for. Building a portfolio of various projects definitely helps sharpening one’s wits, developing one’s own analytical ‘eye’ and experimenting with different methods that can unlock new ways of thinking about the data and possible solutions. Solutions that don’t need to be perfect but are ready to be tested and refined.

One common way to develop insights during ‘quick and dirty’ analysis is narrowing down our material to a few exemplary statements from our informants. These could be statements that truly stood out or that were perhaps repeated by several of the people we talked to and are particolarly relevant to our research question(s). Finding the good statements often calls for skimming the inevitable cliches from the more relevant bits. The process of picking the good stuff comes with practice, a very open mind and a very critical approach to the field.

And how to develop design ideas from these insights? Well the answer is, again, creativity and imagination — aided by some good methods like the one in this ebook for example. And lots of practice. The goal is to generate novel ideas, in a process a bit similar to that of jazz improvisation or MC freestyling.

A good analytical process can produce a lot of material that can be rearranged and recombined in new ways.

Tagged on: