Evaluating the Graph-based Visualization Technique: A Controlled Experiment

Many researchers have highlighted the scarcity of empirical studies that systematically examine the advantages and disadvantages of the use of visualization techniques for software understanding activities. Such studies are crucial for gathering and analyzing objective and quantifiable evidence about the usefulness of proposed visualization techniques and tools, and ultimately, for guiding the research in software visualization. This paper presents a controlled experiment aimed at assessing the impact of a graph-based visualization technique on comprehension tasks. Six common comprehension tasks were performed by 20 undergraduate software engineering students. The completion time and the accuracy of the participants’ responses were measured. The results indicate that on one hand the use of the graph-based visualization increases the correctness (by 21.45% in average) but on the other hand it does not reduce the completion time in program comprehension tasks.