Principles of Graph Construction

  1. Graphical Perception

  2. Principles

GRAPHICAL PERCEPTION

Goal: Communicate information

Author (you) -> graphic -> Reader

A graph is more effective than another if its quantitative information can be decoded more quickly or more easily by most observers.

Nancy Robbins, Creating More Effective Graphs

Pattern perception depends on

Detection: recognition of geometry encoding values

Assembly: grouping of detected symbol elements

Estimation: assessment of relative magnitude of values

A hierarchy of human graphical perception abilities

  1. Position along a common scale (most accurate task)

  2. Position along identical nonaligned scales

  3. Length

  4. Angle and slope

  5. Area

  6. Volume

  7. Color: hue (red, green, blue, etc.), saturation (pale/deep), and lightness. Hue can give good discrimination but poor ordering.

PRINCIPLES

  1. Show the data

  2. Avoid distorting the data

  3. Aid comparisons

  4. Maximise data:ink

  5. Label data, minimize legends

  6. Use clearly different symbols, colours

  7. Consider 0

  8. Dot plots are often better than bar charts

1. Show the data

plot of chunk unnamed-chunk-2

plot of chunk unnamed-chunk-3

2. Avoid disorting the data

plot of chunk unnamed-chunk-4

plot of chunk unnamed-chunk-5

3. Aid comparisons

plot of chunk unnamed-chunk-7

plot of chunk unnamed-chunk-8

4. Maximise data:ink

plot of chunk unnamed-chunk-9

5. Label data, minimize legends

plot of chunk unnamed-chunk-10

6. Use clearly different symbols, colours

7. Consider 0

plot of chunk unnamed-chunk-11

plot of chunk unnamed-chunk-12

8. Dot plots are often better than bar charts

Bar charts:

  • High ink:data ratio

  • Error bars cause perception errors

  • Can only show one-sided confidence intervals well

  • Thick bars reduce the number of categories that can be shown

  • Labels on vertical bar charts are difficult to read

Less is More