Graphical perception is the visual decoding of the quantitative and qualitative information encoded on graphs.1 In the world of data presentation and visualization, the value of graphical perception plays a huge role in the accurate comprehension of analysis presented. The theory of graphical perception provides us with a scientific foundation to apply as framework to even basic perceptual tasks and charts such as bar graphs, pie charts, maps etc.
The underlying principle is that as one encodes information into graphics, the viewer similarly has to decode and interpret results.

2 William Cleveland and Robert McGill lay the foundational theory for setting guidelines on graph construction giving us some best practices to follow:
* Use synchronized scales – it is hard to compare across offset scales
* Compare positions on identical scales when possible
* Limit the use of length comparisons – proportions are difficult to interpret visually without further consideration
* Limit use of pie charts – angular and curvature comparisons are hard to interpret and prone to error
* Do not use 3-D charts or shading – it will skew perception and interpretation of information

setwd("~/Desktop/HU/512/hw4")
knitr::include_graphics("graphical perception essay - image 1.png")

Figure 1: Elementary Perceptual Tasks

The experiments conducted and validated in Crowdsourcing Graphical Perception 3 using Mechanical Turk also teach us the value and efficiencies in this method to gain additional insights through data visualization and experimentation.

Another research study 4 conducted by C. Melody Carswell to predict performance of graphical displays as a function of the particular specifier or visual dimension. The study found that minimal differences were obtained among graphs that used position, length, or angle as specifiers and graphs that used area or volume to represent quantitative values were associated with consistently worse performance than other formats.

In conclusion we can draw that this vast vertical of graphical perception is an essential area of study to accurately present and comprehend information. The fundamentals covered in the papers referred as well as experimental procedures are a guided framework to minimize ambiguity and errors in the way we visualize data.

References:

1 William S. Cleveland, Robert McGill: Graphical Perception and Graphical Methods for Analyzing Scientific Data American Association for the Advancement of Science Vol. 229, No. 4716, pp. 828-833, 3Aug 1985
2 William S. Cleveland, Robert McGill: Graphical Perception: Theory, Experimentation, and Application to the development of graphical method Journal of the American Statistical Association Vol. 79, No. 387, pp. 531-554, Sep 1984
3 Jeffrey Heer, Michael Bostock: Crowdsourcing Graphical Perception Using Mechanical Turk to Assess Visualization Design ACM Human Factors in Computing Systems (CHI), 203-212, 2010
4 C. Melody Carswell: Choosing Specifiers: An Evaluation of the Basic Tasks Model of Graphical Perception Vol. 34, No. 5, pp. 535-554, Oct 1992