Investigating how aesthetics can enhance people’s perceptual awareness of uncertainty in line graph data visualisations
This research aims to enhance people’s perception of uncertainty in line graph data visualisations. It will achieve this through applying aesthetics to line graph data visualisations. The author of this thesis suggests that aesthetics can promote a heightened perceptual awareness of uncertainty in line graph data visualisations and extend the more conventional uncertainty presentation techniques to offer alternative visualisation methods.
The research is reflected through three studies (1) empathising with the end users both in the fields of data science and design, in which 10 (data science) and 4 (design) semi-structured interviews were conducted. (2) understanding the most influential aesthetic renderings, in which 1142 fully completed mixed-methods questionnaires were collected. (3) comparative and user tasks analysis, in which a range of aesthetic renderings (high, medium and low uncertainty) were evaluated through an art critique approach with 531 participants in the context of real-world scenarios.
The findings highlight the power aesthetic renderings of uncertainty have on encouraging a lay audience’s ability to have a heightened sense of uncertainty through the visual depictions of data. Certain aesthetic depictions were shown to create a heightened awareness of uncertainty in line graphs, causing participants to respond more cautiously to a visualisation. By determining what aesthetic renderings afford different levels of uncertainty, the research opens alternative aesthetic visualisation methods for uncertainty visualisation. In an era of increased data usage, the intuitive visual representation of uncertainty is vital to creating a transparent and human-centred approach to variability in data.
Knowledge Economy Skills Scholarship (KESS2)
- School of Technologies