Visualizing Streaming of Ordinal Big Data

Horizontal transitions are used in the Streaming of Big Data when there is the need to change the aggregation level of the data being presented. For example, data in a heat map may be aggregated into a line chart. Although these transitions have already been studied for quantitative streamed big data, ordinal data remains unchecked. In this study, we conducted an empirical study to explore horizontal transitions for ordinal data using Graceful Degradation, a concept that allows an overview of the received data at different periods via different levels of aggregation. We chose four visual idioms (Histogram, Ordinal Scatter Plot, Heat Map, and Line Chart), created several transitions between them, and tested how effectively could people perceive data in each idiom before, during, and after each corresponding transition. Participants had to watch numerous videos showcasing the idioms and transitions, and then they had to answer a questionnaire for us to measure how effective was their perception. All the four idioms tested were effective, and we were able to define numerous design guidelines for the creation of horizontal transitions in Streaming of Ordinal Big Data.