Evaluation of storytelling in information visualization (MPhil to PhD Transfer Report)

Badawood, D. (2012). Evaluation of storytelling in information visualization (MPhil to PhD Transfer Report). London, UK: City University London.

[img]
Preview
PDF
Download (530kB) | Preview

Abstract

Story telling has been used throughout the ages as a means of communication between people and to convey and transmit knowledge from one person to another, and from one generation to the next. In various domains, formulating of messages, ideas, or findings into a story has proven its efficiency in making them understandable, comprehensible, memorable, interesting, and engaging. Information Visualization as an academic field has also utilised the power of storytelling to make visualizations more understandable and interesting for a variety of audiences, including experts. However, although storytelling has been a hot topic in information visualization for some time, little or no empirical evaluations exist to compare different approaches of storytelling through information visualization. There is also a need for work that addresses in depth some particular criteria and techniques of storytelling such as transitions types in visual stories in general and data-driven stories in particular.

A within subject experiment with 13 participants has been conducted to explore empirically how two different models of story delivery with information visualization influence narratives/stories constructed by audiences. Specifically, the first model involves direct narrative by a speaker using a visualization design to tell a story, while the second model involves constructing a story by interactively exploring visualization software. An openended questionnaire in controlled laboratory settings has been used in which the primary goal was to collect a number of stories derived from the two models. All the stories written by the participants were transcribed, analysed, and coded, using data-driven and preset themes. Themes included initial perception of the main story pattern/topic, insight types derived, narrative structures, and unexpected type of insights gained. This experiment was followed by a semi-structured interview where each participant answered two Likert-scale questions on each delivery model, and commented on the overall experiment. It is found that although most participants found telling a story easier with the first model (narrative) they did not perform better in other aspects. The second model (software) was advantegeous in the variety of insight types gained and participants accepted the message and information more neutrally. In contrast, participants were more critical about the data in software model than in the narrative model. The role of time in structuring story events was more apparent in the software model. These findings have some significant practical implications on storytelling through information visualization. A statement of the work done and a work plan for the remaining period of the PhD is also included explaining the proposed enhancement to the experiment conducted and further research work planned to address the issue of transitions in storytelling visualization.

Item Type: Monograph (Other)
Subjects: Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Divisions: School of Informatics > giCentre
City University London PhD theses
URI: http://openaccess.city.ac.uk/id/eprint/1741

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics