Updated: Dec 15, 2020
The ability to exchange analysed qualitative data between CAQDAS-packages with the REFI-QDA Standard revolutionises opportunities for teaching computer-assisted qualitative analysis. In this post, I explain why.
The REFI-QDA Standard - what it is and where to find out more
Developers of six leading CAQDAS-packages (ATLAS.ti, f4analyse, NVivo, QDA Miner, Quirkos, and Transana) have been working together since September 2016 to develop a free and open-source standard for exchanging analysed qualitative data between products. It is called the REFI-QDA Standard (REFI stands for the Rotterdam Exchange Format Initiative).
I've had the privilege of co-coordinating this endeavour with several colleagues* (see below for full details). The exchange standard has many benefits - outlined on the project website, and in previous posts by myself and Daniel Turner. Here, I share some of my recent experiences exchanging analysed data between CAQDAS-packages using it.
I spend much of my time helping students and researchers at various stages of their careers, working in different disciplines and sectors, using different methodologies and with different outcome requirements. I love this work, especially familiarising myself with diverse projects, and working out expedient and creative ways to facilitate the accomplishment of each project’s analytic needs, within their methodological and practical contexts.
Two of the sessions I often deliver benefit from exchanging analysed qualitative data between products: comparative software planning/showcasing workshops, and software-specific training sessions.
Showcasing the contrasts between CAQDAS-packages
The REFI-QDA Standard helps me showcase the affordances of and differences between CAQDAS-packages because I can more easily illustrate and discuss the same research project in different CAQDAS-packages. This makes it easier to concretely demonstrate the distinguishing features of each program, as well as the sometimes subtle, and sometimes profound, differences between them.
Up until now, I have usually demonstrated different research projects in different CAQDAS-packages – either because the examples were drawn from real-world projects I’ve been involved in (that were accomplished in one or other CAQDAS-package), or because they were the sample projects provided by the software developers. I have also manually re-created some of my examples in other CAQDAS-packages, but this is time-consuming, and every time I make a change in one, I have to do the same in all the others.
Now that I can convert an analysis originally conducted in one CAQDAS-package into another I can show how the same analysis would be undertaken using the tools provided by each program without having to recreate the projects. This gives me a wider range of examples to draw from and allows me to demonstrate contrasts between products in a broader range of analytic situations. Although some research examples require the features of one particular program, overall the opportunities for direct and concrete comparisons between packages using the same research example have been greatly extended by the REFI-QDA Standard.
From the researcher’s point of view, some value seeing how a research project that is closely aligned with theirs, in terms of data-types, analytic strategies, and outcome requirements, could be undertaken in different programs. This allows them to focus attention on the idiosyncrasies of the tools, rather than having to translate a research example from a different research tradition into their own methodological context.
Additionally, researchers who are unfamiliar with CAQDAS-packages appreciate seeing different types of analysis projects in several different CAQDAS-packages, to raise their awareness in general terms. And others may be looking to select a CAQDAS-package for a variety of longer-term needs, and want to see as many different research applications and software options as possible.
The REFI-QDA Standard therefore allows me to more easily and efficiently meet all these needs when showcasing the affordances of and differences between CAQDAS-packages.
Illustrating the flexibility of CAQDAS-packages
When running training sessions focusing on a single CAQDAS-package I always show several different research projects to illustrate the full range of software features, the flexibility of the program, and the importance of ensuring that analytic strategies drive the use of software tools.
However, many CAQDAS-packages now include dozens of tools, and learners of these programs sometimes imagine they won’t be using it ‘properly’ or to its ‘full potential’ if they don’t use every available tool. I counter this common myth by reminding them that they don’t use every available tool in, for example, Microsoft Word or Excel, yet they still successfully accomplish their writing or spreadsheet tasks.
It’s important to show several different research projects within one CAQDAS-package for two main reasons:
to demonstrate the extensiveness of the program. Even though they might not need all the program’s tools for any particular project, learners need to understand everything about the program at a high-level in order to make informed decisions about which tools to use to accomplish particular analytic tasks.
to illustrate the importance of harnessing the chosen program appropriately for the unique needs of each analysis. This principle is central to my teaching (and fully explained in the Five-Level QDA method books that I wrote with Nick Woolf). There is no ‘correct’ or ‘ideal’ way to use any CAQDAS-package, because every analysis has its own idiosyncrasies. It’s therefore important to show how the same tools can be harnessed differently in different analysis projects and to show how the same analytic task can be accomplished using different software tools.
But it’s not just when teaching the larger CAQDAS-packages that have dozens of tools that this is important. When teaching CAQDAS-packages with fewer tools, it’s also essential to show different research projects in the one program, because these programs are also flexible and their fewer tools can also be used for a variety of purposes. More is not necessarily better, just as fewer is not necessarily restrictive. That’s one of the delights of computer-assisted qualitative data analysis.
My long-established teaching methods are greatly eased by the REFI-QDA Standard, as the wide range of analysis projects I originated in any one program can be converted to other programs. This extends the possibilities for illustrating both the tools and the flexibility of the programs, and for demonstrating creative ways of harnessing the tools of multiple programs in many different types of analysis.
* I am currently co-coordinating REFI with Kristi Jackson (Queri Inc.), Graham Gibbs (University of Huddersfield), and Fred Van Blommestein (University of Groningen). Previous co-coordinators were Jeanine Evers (Evers Research & Training), initiator of the endeavour, Elias Rizkallah (Université du Quebec à Montreal), and Yves Marcoux (University of Montreal).