Xiaole Kuang, Haimo Zhang, Shengdong Zhao, Michael J. McGuffin
One of the fundamental tasks for analytic activity is retrieving (i.e., reading) the value of a particular quantity in an information visualization. However, few previous studies have compared user performance in such value retrieval tasks for different visualizations. We present an experimental comparison of user performance (time and error distance) across four multivariate data visualizations. Three variants of scatterplot (SCP) visualizations, namely SCPs with common vertical axes (SCP-common), SCPs with a staircase layout (SCP-staircase), and SCPs with rotated axes between neighboring cells (SCP-rotated), and a baseline parallel coordinate plots (PCP) were compared. Results show that the baseline PCP is better than SCP-rotated and SCP-staircase under all conditions, while the difference between SCP-common and PCP depends on the dimensionality and density of the dataset. PCP shows advantages over SCP-common when the dimensionality and density of the dataset are low, but SCP-common eventually outperforms PCP as data dimensionality and density increase. The results suggest guidelines for the use of SCPs and PCPs that can benefit future researchers and practitioners.
Additional information:
Scatter plot is an information visualization technique that can present multivariate data and claimed to be useful in revealing relationship between two dimensions in the data. Another technique called parallel coordinate plot is inherently designed to have the advantage of being able to trace a particular data tuple across multiple (>2) dimensions. In this project, we aim to investigate the advantage and differences between the two visualization techniques. Specifically for the current phase, we focus on value retrieval tasks, in which the user is required to determine the value of a data tuple along one dimension given its value on another dimension. Our methodology is mainly based on GOMS (Goal, Operator, Method, and Selection criteria) method, in which we try to determine the basic operators required to achieve the task and their cost in terms of performance time. We hope to reveal the advantage of PCP over SP in a systematic approach that guides future information visualization applications in selecting the best techniques according to the specific tasks.