Graphical Perception Essay

In 1984, William Cleveland and Robert McGill [1] published a seminal study on the elementary perceptual tasks which are the most basic visual tasks we perform in our perception of graphs.

# Cleveland-McGill Elementary Perceptual Tasks
knitr::include_graphics("C:/Harrisburg University Classes/Sem 2 - Fall 2017/ANLY-512 Data Visualisation/Assignments/Perception Essay/ElementaryPerceptualTasks.png")

According to Cleveland and McGill [1], the easiest perceptual task is to “compare position along a common scale”. Scatterplots and bar charts are good examples of comparing objects on a common scale. However, this perceptual task can be extended to judgements of length and area. In 1910, Heer and Bostock [2] revisited some of the Cleveland and McGill’s old experiments in a study they did as a proof of concept for using Mechanical Turk users in scientific studies and found similar results as those of Cleveland and McGill.

# Cleveland-McGill & Heer-Bostock Elementary Perceptual Tasks
knitr::include_graphics("C:/Harrisburg University Classes/Sem 2 - Fall 2017/ANLY-512 Data Visualisation/Assignments/Perception Essay/ClevelandHeerProportionalJudgements.png")

There are a variety of other graph elements that helps define our perceptual judgements. Here, we take a look at reference points, basic shapes and bar vs pie charts for proportions.

Reference Points: Schiano and Tversky [3] found that there is a systematic bias towards an imaginary 45-degree line in line charts. Their work suggests that different visual systems promote different visual frames.

Basic Shapes: Croxton [4] had shown about 8 decades ago that bars were he most effective in communicating comparative values than either circles, squares or cubes with cubes being the worst.

Bars and pies for proportions: Simkin and Hastie [5] had participants make proportional judgements and segment-to-segment (comparison) judgements. They found that for segment-to-segment judgements, simple bar charts worked best, followed by divided bar charts and then pie charts. However, for making proportional judgements the reverse trend is true with pie charts being the best option.

# Simkin-Hastie Segment-to-Segment and Proportional Judgements
knitr::include_graphics("C:/Harrisburg University Classes/Sem 2 - Fall 2017/ANLY-512 Data Visualisation/Assignments/Perception Essay/SimkinHastieSegmentProportional.png")

References

  1. Graphical Perception: Theory, Experimentation, and the Application to the Development of Graphical Methods. William Cleveland and Robert McGill, 1984.

  2. Crowdsourcing Graphical Perception: Using Mechanical Turk to Assess Vissualization Design. Jeffrey Heer and Michael Bostock, 2010.

  3. Structure and Strategy in Encoding Simplified Graphs. Diane Schiano and Barbara Tversky, 1992.

  4. Graphical Comparisons by Bars, Squares, Circles and Cubes.

  5. An Information-Processing Analysis of Graph Perception. David Simkin and Reid Hastie, 1987.