Block 2: Week 7 Visualisation
How am I engaged?
This week I decided to collect data on the how of my engagement in all my work and study projects.
Methodology
For five days I recorded all my activity (Eynon’s (2015) ‘what’), and categorised it by how I engaged: watching, listening, reading, considering, speaking and writing. I also noted if I was working alone or in a group.
Results and Analysis
I decided this time to focus on creating an easy-to-read visualisation; it might be thought that to a teacher, I’m just part of a large cohort, so there wouldn’t be much time to view my individual progress.
Since I was analysing activity, rather than using time, I simplified the data by grouping it together by the ‘project’ (one line of the visualisation for each), then reduced its representation to the minimum possible (a single letter or symbol). This reduction was to enable quick visual analysis.
Furthermore, to enable a quick valuation of the data, I deliberately used a representation to suggest a judgement of how I was engaged in each project.
This is inspired by the ideas of negative space and that what is not there can get more attention that what is: what are your eyes drawn to in the following image and what assumptions are behind that?
If a dashboard shows you what is not present, does that imply it is ‘missing’ and that it should be there? Would you have looked for it, if it had not been indicated? What is missing and not acknowledged?
Learning activities are already loaded with value (e.g. passive or active) by the individual. What if a dashboard suggests some data are more valuable than others, but doesn’t make its underlying bias as obvious as my visualisation; how much could be accepted without question?
How does this relate to teaching?
- It might be suggested that a dashboard makes a complex situation easy to understand. However, such a property is commonly accepted as only positive whereas, in this case, simplification is at a cost.
- What a teacher sees on a dashboard, will be determined by what the learning system has already been programmed to record, and what of that is shared. The teacher may be in a position to select from that but may still have to accept how it is displayed, which in some way transforms it. There is, at best, an illusion of teacher empowerment, but that may be no more than choice over pre-selected options.
- There will, of course, be data that cannot be recorded, and any data that is self-reported may be valued less highly (Williamson et al, 2020). That a dashboard differentiates between the value (trustworthiness?) of data in a display, could be suggestive in itself. Again, this may be outside the teacher’s control.
- If this data is visible by the institution, this may impact how a teacher views it: a proxy for their teaching, rather than about student activity (Williamson et al, 2020) and a means whereby they are datafied (Williamson, 2016). This may impact their teaching as, if they feel it reflects on them, they may make changes in order to ‘improve’ the dashboard, rather than for pedagogic reasons (Brown, 2020). They may also experience this as surveillance, disempowering and diminishing any autonomy they have.
- Dataist comparisons between teachers (Williamson et al, 2020) may not make sense, but may be enabled by the learning system itself. That a dashboard is made so simple, is applied across all contexts, with data collected and centralised, could be said to invite comparison. So, as strategies for students are identified through their datafication, teachers might find themselves nudged by predictive analytics.
References
Brown, M., 2020. Seeing students at scale: how faculty in large lecture courses act upon learning analytics dashboard data. Teaching in Higher Education. 25(4), pp. 384-400.
Eynon, R., 2015. The quantified self for learning: critical questions for education. Learning, Media and Technology, 40 (4), pp. 407-411, DOI: 10.1080/17439884.2015.1100797
Williamson, B. 2016. Digital Education Governance: Data Visualization, Predictive Analytics, and ‘Real-Time’ Policy Instruments. Journal of Education Policy. 31 (2), pp. 123–141. doi:10.1080/02680939.2015.1035758
Williamson, B. Bayne, S. Shay, S. 2020. The datafication of teaching in Higher Education: critical issues and perspectives. Teaching in Higher Education. 25(4), pp. 351-365.
“…to enable a quick valuation of the data, I deliberately used a representation to suggest a judgement of how I was engaged…” I thought at first that “valuation” was a mistake and you meant “visualization” but I believe this is intentional? The question of how visualizations perform a kind of valuation seems really significant, especially in relation to your comments about things that are (not) made visible from data. (“Negative space” is a new one for me, so thanks for that too!) What is made visible appears to have value, and what’s left invisible is presumed value-less? But as you note, “What if a dashboard suggests some data are more valuable than others, but doesn’t make its underlying bias as obvious as my visualisation; how much could be accepted without question?” This seems to me an essential question. And it leads to the issue that some aspects of learning might be amenable to being captured and represented as data–e.g. activities, behaviours, and other “observable” features–but many others aren’t–critical thinking, reasoning, making judgments, prior beliefs, biases and assumptions that colour an interpretation etc. By being visualized, do the former observable quantities attain greater “value” over others that can’t be visualized? Does it narrow the field of perception of the educator to the single output of an algorithm rather than the complex multiplicity of education and learning? Your visualizations and commentaries demonstrate your very careful and thoughtful engagement with such difficult issues and questions.
Yes, it was a typo but for ‘evaluation’, so yes, I did mean to imply that the viewer looks at the visualisation with a view to determine its ‘worth’. It seems to be implicit that if we turn data into a dashboard (or another device, like an infographic) then it must be of use, all we have to do is determine what that use is. We appear to have dispensed with the first question which is ‘Can data such as this ever tell me anything I wish to know or could make use of?’