Evaluating the Influence of Surprise and Suppression Techniques in Map Visualizations

Ndlovu, Akim

Etd

Evaluating the Influence of Surprise and Suppression Techniques in Map Visualizations

Public Deposited

Choropleth maps are commonly used to visualize geospatial data, such as disease outbreaks across various geographic regions. However, well-known biases associated with choropleth maps, such as the effect of area in map exploration tasks and population statistics in the interpretation of event rates, have led to extensive research on how to overcome such biases to avoid misleading users. Two recently developed techniques, Surprise and VSUPs (Value Suppressing Uncertainty Palettes) may be considered as viable solutions for overcoming biases in choropleth maps, but have yet to be empirically tested with users of visualizations. In this thesis, we explore how well people make use of Surprise and VSUPs in map exploration tasks, by conducting a crowdsourced experiment where n = 300 participants are assigned to one of Choropleth, Surprise (only), and VSUP conditions (depicting rates and Surprise in a suppressed palette). We improve participants’ exploration of each stimulus through the use of various interaction techniques (e.g. zooming and panning), and adapt tasks from prior studies to reduce noise from participants’ responses. Quantitative analysis shows clear differences in the interpretation of metrics such as rate, surprise and population, with surprise maps leading people to map locations with significantly high population and VSUPs performing similar or better than Choropleths for rate selection. In addition, qualitative analysis suggests that many participants may only consider a subset of the metrics presented to them during map exploration and decision-making. We discuss how these results generally support the use of Surprise and VSUP techniques in practice, and opportunities for further technique development. For replicability and reproducibility, the material for the study (data, study results and code) is publicly available at https://osf.io/exb95/.

Creator