Evaluating trace link visualizations

Abstract

Traceability has become a very important part of the software development lifecycle and therefor a lot of research has been done and is being done about it. Our aim with this paper is to explore and evaluate the effect that different traceability visualizations have on the workload and find the most optimal visualizations that users will benefit more by using in their projects. We used design science research in this paper and started with a survey of industry professionals so that we could get an overview of traceability and visualizations use in real life. This information was then compared with the results of the experiment that was conducted as a second step. We created prototype visualizations that were used for the experiment in which 16 students took part. These two iterations provided us with qualitative and quantitative results that showed us there is no difference in workload between the visualizations and that perception and familiarity have an influence in which visualization is preferred. Furthermore the results indicated that no matter the visualization the information should be provided by them is the same.