Visualizing answers to support discussion in group-learning environments

Abstract

In this master thesis we discuss challenges in creating visualizations with the goal of enhancing discussions in group learning environments. We focus on two types of task to visualize. The first task type involves visual presentation of free-text related to the discussion in groups, and the second task type involves ranking of certain attributes on different objects. The free-text needs to be filtered in the effort to support shared meaning and discussion. This entails structuring text data in order to derive information and represent answers both on an individual and a more holistic level. We explore how to best visualize answers by utilizing animated transitions and passive interactivity. This thesis explains how the visualizations were conceptualized, prototyped and evaluated. By utilizing a scrum based design process, we try to explore different possibilities to represent our data and learn about the degree of complexity that is suitable for the intended users of these visualizations. The summative evaluations demonstrate the usefulness of text processing algorithms and immersive experiences in order to scaffold and drive rich discussions.