Empirical Investigation of Typographic Overprinting Displays and Their Legibility in the Context of Information and Geographic Visualization of Text

Open Access
- Author:
- Savelyev, Alexander
- Graduate Program:
- Geography
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 10, 2015
- Committee Members:
- Alan Maceachren, Dissertation Advisor/Co-Advisor
Alan Maceachren, Committee Chair/Co-Chair
Donna Jean Peuquet, Committee Member
Alexander Klippel, Committee Member
Prasenjit Mitra, Committee Member
Cynthia Ann Brewer, Special Member - Keywords:
- cartography
typography
legibility
geovisualization
information visualization
overprinting - Abstract:
- This dissertation investigates the visual technique of typographic overprinting – printing one layer of text directly on top of another, allowing words and letters to overlap. First, a survey of geovisualization and information visualization literature demonstrates potential application areas for this technique. Second, a novel methodology is proposed that can be used to assess the legibility of information visualization displays built using the typographic overprinting technique. This methodology is synthesized based on a survey of information visualization, cartographic, typographic and reading research. Third, two human subject experiments are described that were designed to validate the proposed methodology and explore the legibility of the typographic overprinting technique displays, respectively. The first experiment, performed on Amazon Mechanical Turk with 104 participants, was used to validate the critical aspects of the methodology in question – the cognitive task, the reading material, and the performance metric. The second experiment, performed in laboratory conditions with 36 participants, was used to explore the connection between the legibility of the typographic overprinting technique displays and two visual variables that were hypothesized to affect said legibility – size and brightness ratio of the two layers of text displayed. Finally, the results of the two experiments are presented, analyzed and discussed. Both of the experiments fulfill their original purpose, as demonstrated by the statistical analysis and the associated p-values, effect sizes and statistical power, and provide multiple leads for future investigations in cartography, geographic and information visualization communities.