International Conference Information Visualisation pages:637-648
IV'07 location:Zurich, Switzerland date:2-6 July 2007
Aesthetic seems currently under represented in most current data visualization evaluation methodologies. This paper investigates the results of an online survey of 285 participants, measuring both perceived aesthetic as well as the efficiency and effectiveness of retrieval tasks across a set of 11 different data visualization techniques. The data visualizations represent an identical hierarchical dataset, which has been normalized in terms of color, typography and layout balance. This study measured parameters such as speed of completion, accuracy rate, task abandonment and latency of erroneous response. Our findings demonstrate a correlation between latency in task abandonment and erroneous response time in relation to visualization's perceived aesthetic. These results support the need for an increased recognition for aesthetic in the typical evaluation process of data visualization techniques.