Data visualization is an effective way to improve the decision making process by using the statistic tools like line chart, bar chart and pie chart to display the data. Humans are able to understand trends, patterns, and relationships between the data from a visualization, and make their own judgement about the data.
Graphical perception is the visual decoding of information encoded on graphs, according to the recent research paper written by Cleveland and McGill. Cleveland and McGill also done some experiments on how people’s ability to perceive the displayed data from varies types of graphs. They provide 10 elementary perceptual tasks that people usually use to interpret from the graphs as follows:
Elementary perceptual tasks
Another finding from Cleveland and McGill’s study is that the order of human’s judgement accuracy, from most accurate to least.
1 Position along a common scale.
2. Position along nonaligned scales.
3. Length, directions, angle.
4. Area.
5. Volume, curvature.
6. Shading, color saturation.
The second paper by Heer and Bostock is about the crowdsourcing graphical perception, they conducted their Amazon’s Mechanical Turk experiments based on the findings of Cleveland and McGill. They reached the same result as Cleveland and McGill.
Stephen emphasizes in his paper that data visualization is to present information graphically for two purposes: sense-making(also called data analysis) and communication. People can use visualization to translate abstract information into physical vision (length, position, size, shape and color), this can only succeed if people understand the perception. He also points out that data visualization can not only represent quantitative values, it can also display characteristic relationship. Eugene and Arnab’s paper explore deeply on how data visualization can make communication and decision making better.
Death by diseases
Graphical perception is the base of the visualization, it can help people understand the data and make a better decision.