Path tracing is a common task in many real world uses of graphs that display networks of relationships. Despite previous work in the evaluation of how factors, such as edge-edge crossings, impact the readability of graph layouts, what makes one path-tracing task more difficult than another is not well understood.
For my M.Sc thesis, I conducted an observational user study with 12 participants completing a path-tracing task. I developed a series of visualizations to support an extensive qualitative analysis of the study data, which led to a detailed characterization of common path-tracing behaviours. I then created a predictive model of the paths that users are most likely to search, named the search set, based on the behaviours that I observed and characterized. To validate the predictive behavioural model, and to demonstrate how the search set could be used, I also conducted a careful comparison of graph readability factors through a hierarchical multiple regression analysis.
process and methods
visualization design and implementation
qualitative coding analysis
multiple regression analysis
Dawson, J.Q., Munzner, T. and McGrenere, J. (2014). A search-set model of path tracing in graphs. Journal of Information Visualization, In Press.
[project page] [pre-print pdf][sage online]
Dawson, J.Q. (2013). A Search Set Model of Path Tracing in Graphs. M.Sc. Thesis, Dept. of Computer Science, University of British Columbia, October 2013. [pdf]
Thesis presentation given at UBC Computer Science on Oct 10, 2014. [slides]