Sketchy rendering to improve data engagement
Research by Professors Jo Wood and Jason Dykes from the giCentre in Informatics (with colleagues from European universities) has presented a framework for constructing 'sketchy' information visualisations - such as maps, diagrams and charts - in a way that mimics data graphics drawn by hand.
The framework is designed to be easily integrated into existing computer-based visualisation platforms with minimal programming modification or effort. The research demonstrates and evaluates examples of statistical maps and graphics conveyed in a sketchy style, to demonstrate spatial imprecision and to enhance the aesthetic and narrative qualities of visualisation.
The historical use of computing technologies for rapid and precise graphical rendering of data means that viewers are accustomed to an objective and authoritative style of visualisation. The researchers propose an alternative style that mimics hand-drawn graphics to imply human input into the design, which is intended to improve engagement while maintaining the advantages of using computers to process complex data.
Evaluations of the framework, published in IEEE Transactions on Visualization and Computer Graphics and to be presented at the world's major visualisation conference in October, show that ability to estimate areas or ratios in sketchy visualisations of data is impaired as sketchiness increases. However, users show improved engagement and more positive participation with the visualisations if annotation is present.
In their paper, Professors Wood and Dykes explain:
The results of our work have implications for effective information visualisation design that go beyond the traditional role of sketching as a tool for prototyping or its use for an indication of general uncertainty.