1. Graphical Excellence
Graphical Excellence is the well-designed presentation of interesting data through
substance, statistics, and design.
Give the viewer the greatest number of ideas
in the shortest time
with the least ink.
Excellent graphics communicate complex ideas with:
- Clarity
- Precision
- Efficiency
🧠 Remember:
Great Visualization =
More Ideas + Less Ink
2. Achieving Graphical Excellence
- Tell the truth about the data (Graphical Integrity)
- Visualize with clarity and precision (Design Principles)
Exam Keyword:
Graphical Excellence = Integrity + Design
3. Graphical Integrity
Graphical Integrity means presenting data honestly without misleading viewers.
"Not lying with statistics."
The purpose is to tell the truth about the data.
4. Principle 1: Avoid Distortion (Lie Factor)
The visual representation of data should be directly proportional
to the actual numerical values.
Lie Factor measures graphical distortion.
Lie Factor =
Size of effect shown in graphic
÷
Size of effect in data
Accurate Graph:
Lie Factor = 1.0
🧠 If Lie Factor > 1
Effect is exaggerated.
5. Principle 2: Avoid Ambiguity
Use clear, detailed, and thorough labels.
- Label axes clearly
- Explain data directly on charts
- Annotate important events
Good labeling prevents misunderstanding.
6. Principle 3: Show Data Variation, Not Design Variation
Emphasize changes in data, not decorative chart effects.
Fancy graphics can create patterns that do not actually exist.
Viewers should notice data changes,
not design tricks.
Avoid unnecessary decorations and visual effects.
7. Principle 4: Account for Inflation
Monetary values should be adjusted for inflation when comparing across time.
RM100 in 2000 does not have the same value as RM100 today.
Adjusted values provide more accurate comparisons.
8. Principle 5: Match Dimensions
The number of visual dimensions should not exceed the number of dimensions in the data.
Avoid using 3D bars to represent simple 1D values.
Extra dimensions often distort perception.
9. Principle 6: Show Data in Context
Graphics should never quote data out of context.
Showing only selected years can create misleading trends.
Context helps viewers interpret data correctly.
10. Tufte's Six Principles of Graphical Integrity
| Principle |
Main Idea |
| 1 |
Avoid distortion (Lie Factor = 1) |
| 2 |
Use clear labels |
| 3 |
Show data variation, not design variation |
| 4 |
Adjust monetary values for inflation |
| 5 |
Match graphic dimensions to data dimensions |
| 6 |
Provide context |
11. Data-Ink Maximization
Data-Ink Ratio measures how much ink is used to display actual data.
Data-Ink Ratio =
Data Ink
÷
Total Ink Used
Maximize data ink and minimize unnecessary ink.
12. Tufte's Five Laws of Data-Ink
- Above all else, show the data
- Maximize the data-ink ratio
- Erase non-data ink
- Erase redundant data ink
- Revise and edit
🧠 Show → Maximize → Erase → Erase → Revise
13. Remove Chartjunk
Chartjunk refers to unnecessary visual elements that do not communicate information.
3D effects, excessive colors, shadows, decorative icons.
Chartjunk distracts viewers from the data.
14. Multi-Functioning Graphical Elements
Every graphical element should perform one or more useful data-related functions.
- Stem-and-leaf plots
- Chernoff Faces
- Coordinate labels as marks
Make every visual element contribute information.
15. Data Density
Data Density measures how much information is displayed in a given graphical area.
Data Density =
Number of Data Entries
÷
Area of Graphic
More useful information can be displayed without increasing chart size.
16. Small Multiples & Parallel Sequencing
Small Multiples are a series of similar charts placed together for comparison.
Monthly sales charts displayed side-by-side.
Makes comparisons easier while maintaining consistency.
17. Deceptive Charts to Avoid
- Bars that do not start at zero
- 3D charts
- Pie charts with many categories
- Unsynchronized dual axes
- Reversed scales
Exam Tip:
Most misleading charts violate Graphical Integrity principles.
18. Visualization Tool Stack
| Layer |
Examples |
| Charting Tools |
Excel, Google Charts |
| Interactive Exploration |
Tableau, Qlik, Power BI |
| Visual Analysis Grammars |
VizQL, ggplot2, Vega-Lite |
| Visualization Grammars |
D3.js, Vega |
| Component Architectures |
Prefuse, Flare, VTK |
| Graphics APIs |
Python, R, OpenGL, Java2D |
19. The KISS Principle
"It seems that perfection is reached not when there is nothing to add,
but when there is nothing left to take away."
KISS = Keep It Simple, Stupid.
Simplicity improves clarity and understanding.
🧠 Simple charts are usually better charts.
20. Final Exam Summary
Most Important Points
- Graphical Excellence:
Most ideas, shortest time, least ink.
- Lie Factor:
Should equal 1.0.
- Graphical Integrity:
Tell the truth about data.
- Data-Ink Ratio:
Maximize data ink.
- Chartjunk:
Remove unnecessary decorations.
- Data Density:
More information per area.
- Small Multiples:
Multiple similar charts for comparison.
- KISS Principle:
Keep visualizations simple.
- Tool Stack:
Excel → Tableau/Power BI → ggplot2 → D3.js → APIs.