How does qualitative software influence analysis?
I found myself wondering — how does Transana, with its affordances and tools and ways of working, influence researchers’ approach to their data?
At about the same time, I also read I read the Lambie et. al. article referenced on the Articles page. In that article, the authors detail how they approached their data in Transana. While reading it, I found myself thinking things like “interesting, that’s not how I would have thought to do that.” Collections around individuals, not around analytic themes or behaviors? Not my first instinct for an organizational scheme for the analytic portion of the data. Collection Maps instead of Keyword Maps? Allows you to focus on one individual at a time from videos showing multiple individuals. Fantastic. Now I get it. Individual Maps as a way of capturing and describing an individual’s response to an event. My way would have been a bit harder (although you could still do it). And given their organization of the data, they can still do all the thematic stuff I would have used as my base Collection structure with searches. Brilliant.
It got me thinking about different ways of approaching data within Transana. I have to admit, this is actually something I think about pretty regularly. (Who’d have guessed, right?) This time, though, I was trying to think about it from Nick’s perspective. I can’t decide if I completely agree with him, at least as far as Transana goes.
I tried to design Transana to be a flexible tool. Transana provides multiple ways to approach data, multiple paths to analyze data, multiple ways to accomplish the same analytic goals. I was very intentional about this. I recognize that different qualitative methodologies have different requirements, and that different researchers have different preferences and styles of approaching data. I’ve tried to come up with ways of meeting the very different preferences of very different researchers, research methods, and data types.
We support Categorizing, the process of creating Collections, creating Quotes and Clips within them, and sorting them, for people who like their analytic data to be visible, tangible, sortable into “piles”, sortable within those piles. We support Coding, the use of Keywords, for researchers who prefer the more abstract act of coding where you may not even pay attention to where the Clip or Quote lands as long as you can see it in the Keyword Visualization. At the extremes, you can work with media data without a Transcript in Transana, you can work with multiple simultaneous transcripts, or you can work somewhere between those extremes using gisted, verbatim, conversation-analytic, and many other types of Transcripts.
Thinking about the Lambie et. al. paper, I think about the different ways one could organize the data, the different ways one could analyze the data, the different ways those analytic choices would impact how easy or hard it is to investigate different questions. But I also look forward to the next time I bump into Nick and Christina so we can talk about this issue of flexibility in software and in analysis. I think it will be a very interesting conversation.