My first data visualisation represented how a student’s conceptual space could expand in response to provocations made via the use of a range of media. I used it to demonstrate how our learning processes are not just cognitive or social but also laden with emotion. My last visualisation aims to demonstrate how the same data on socio-emotional learning processes can be visualised in a different way and recycled for different purposes as Williamson [2017, 81]. For example, the UK’s National Pupil Database makes available pupil data, including sensitive data, to third party analysis. One can easily imagine this being used in a completely different context by a health insurance broker to assess someone’s mental health and then used to inform decisions related to the cost of their insurance premium.

Williamson [2017, 77] writes that images of data are ‘powerful explanatory and persuasive devices’. I’d add that sometimes the design of the visualisation itself constrains what data are represented. For example, I wanted this sea-like design (influenced by this) to be very simple to contrast with the earlier representation on conceptual space which was more complex. But this made it harder to represent the duration thinking and the platform used, so I excluded data representing them. Given that, as Williamson [2017, 77] notes, data visualisations and displays themselves are also used as policy instruments, this means that the visualisation chosen serve rhetorical purposes and can be used to leave certain data out or make others more salient than they might otherwise be. Furthermore, we might present so-called ‘intimate data’ [Williamson 2017: 82] on, for example, emotional responses, in such an aesthetically charming way that we ignore what’s left out or ignore ethical questions about the appropriateness of gathering intimate data on people’s emotional and mental lives. Data may sometimes be beautiful; that doesn’t make them good. Finally, you might think that the collection of data purporting to represent intimate emotional and mental states, and physical states of a person’s body is a form of data colonialism of a distinctively bodily kind without the meaningful consent of the person represented, suggesting in turn a kind of corporate sense of entitlement to a person’s body that the one represented is not able to resist or even contest since it is ‘concerned with the external appropriation of data on terms that are partly or wholly beyond the control of the person to whom the data relates’ [Couldry and Mejias, 2018; Prinsloo, 2020, 367].