Towards a Taxonomy for Evaluating User Engagement in Information Visualization

Abstract

Nowadays, with the availability of massive amounts of personal data, the role of Information Visualization is getting more and more important. There are a lot of visualization and visual analytics tools to help people to make sense of their personal data in their everyday lives. There is a consensus in the visualization community about the importance of user engagement for personal visual analytics. However, there are many challenges associated with this topic. As of yet, there is no clear and widely accepted definition for user engagement. Consequently and despite some recent efforts, there are no systematic and unified methods for evaluating different aspects of user engagement. In this paper, we bring attention to some fundamental open issues in the context of Information Visualization: What is the definition of engagement? What are the levels of engagement? How to measure engagement? How to improve user engagement level? This research is an initial attempt towards addressing some of these issues. Our main goal is to enable visualization researchers to more accurately measure and evaluate user engagement with visualizations. To this end, we reviewed definitions, measures, and frameworks from various other disciplines and specified the gap in the visualization community. We propose a five level taxonomy for engagement which is deeply inspired by Boloom’s Taxonomy. In addition, we address some important open questions that require further exploration. This paper establishes a groundwork for future research in user engagement with visualizations.