Data-relationships and digital surveillance

This is my PhD. I was working as an educational technologist, running workshops: bring some open-source stats, some digital footprints, and see what people make of them. It started as a way to help learners look at their informal learning. Pretty quickly the conversation stopped being about study habits and became about privacy. People were surprised at how far you could get with freely available tools, and they started joining the dots with whatever had been in the news that week: Snowden, GCHQ, Cambridge Analytica.

Teaching the statistical methods did not engage most people. Technical folk turned up wanting to talk about their own code. Non-technical people asked to extract their own Facebook data and then count phrases in R often bounced before they finished.

So I flipped it. I took a dataset a community already cared about, professional wrestling tag-teams, pulled from Wikipedia, drawn in Gephi, and showed them the picture first. The methods sat nearby for anyone who wanted them. People could read the network because they already knew the wrestlers. Then they started asking how it was made, and some of them started wondering who else might be doing this to data about them.

That is what the thesis argues for: methods anyone can inspect, applied to data a community already cares about, shown to those people rather than to a coding class. A later survey suggested that a vague awareness of tech issues does not change behaviour much, but insight into the methods does. That method is still the useful bit. I am not running the workshops as a programme now.