This week I tracked the conversations I had with people on our course. I chose to represent each conversation – with Colin, Tracey, Ben, and Jeremy – as a butterflyesque creature, playing on the idea of ‘social butterflies’ to emphasise the social dimension to both learning and teaching, and on the idea that conversation itself can be a useful methodology to elicit someone’s experience of learning, or, indeed, of teaching [Harrison et al, 2019].
The data visualisation tells you who I’m talking with, the subject we’re talking about, and the platform (University email, Twitter). It could be used to illustrate that there’s no obvious way for a teacher to interpret this data. X talked with Y on platform P about subject S – but so what? Was it a long conversation? Was it just about something relatively minor? Was there, for example, any argy-bargy in conversation with Ben? (Answer=no!). In addition, conversations are often nuanced, subtle, and ambiguous (even for their participants!) in a way that makes them hard for teachers (or anyone else!) to interpret without a whole a lot of contextual detail, and that detail, of course, is what a teacher would need to make proper sense of the conversations represented in the visualisation, and to properly judge whether or not they were helping or hindering a student’s learning. However, even in a small class of 25 students, it’s not obvious that any teacher would have time to interpret and evaluate more detailed data on every student conversation without adding a considerable amount of work to their workload. And this in turn suggests that the demand that teachers become more data literate in the sense of the definition given by Sander [2020:3] may in fact turn out to be unreasonably demanding given the time available (see here for my further discussion). It’s far from obvious to me that such detailed data interpretation should be part of the role of a teacher in the first place.