Discovering Data-Driven Stories : A Case Study

Sammanfattning: Narrative visualization is a young and emerging field, driven mainly by data journalists. For this reason, most data stories available today are author-driven. However, with the rise of interactive visualizations the possibilities for creating reader-driven stories have become apparent. In this thesis, we present a straigthforward prototype, AsylKoll, built to support the articulation of reader-driven stories about Swedish immigration during 2015. We test its ability to support reader-driven stories by performing two user-studies based on the Think Aloud Method. In particular, we evaluate the prototype along the dimensions of reader engagement and learning. We find that user-centric data and various effects, such as transitions and mouse-overs, have a positive impact on reader engagement. In addition, we find that typical tasks such as extracting extremes and making comparisons are very important for users to gain insight and learn from the data. Foremost, this thesis shows the potential that simple, interactive visualizations have to make people engage and gain insights from data.