Project description

Organizations have to process and understand their internal and external data. In order to do so, many organizations use visualization tools to design graphs. However, when actually designing these visualizations, often questions emerge: How should complex data be visualized to improve decision making? How can cognitive overload of the users be prevented? How flexible or interactive should self-service visualization options be?

In the Eurostars-project 'Dr. Tide: Digital Research Tailored Information Detector' it was explored in cooperation with a German and a Norwegian company how complex data should be visualized. For this purpose, visualization rules derived from the literature were empirically tested with both, students and pilot users from one of the participating companies. At the beginning of the project, the focus was mainly on the design of static visualizations. Towards the end of the project, the focus was on self-service visualizations. Using complex data from the agricultural insurance sector (cultivated varieties, weather, damage caused by drought or hail), it was examined whether and how autonomously varying the presentation format helps users to gain a better understanding of the data. The experiments on the effects of self-creation of visualizations showed that the complexity of the question and the previous knowledge of the users play an important role. Overall, the experiments showed that it is useful for Dr. TIDE to provide the users with a limited number of meaningful interaction possibilities.

Kooperationen

Genillard & Co

OSINT Analytics

Publications

Kopp, T., Riekert, M., & Utz, S. (2018). When cognitive fit outweighs cognitive load: Redundant data labels in charts increase accuracy and speed of information extraction. Computers in Human Behavior, 86, 367-376. https://dx.doi.org/10.1016/j.chb.2018.04.037.