CDS 6324 - Data Visualization

Lecture 5: More Visualization Design & Chart Types

1. Revision: Graphical Excellence

Graphical excellence gives the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space.
Excellent visualizations maximize understanding while minimizing clutter.
🧠 More Ideas + Less Ink = Better Visualization

2. Revision: Graphical Integrity

Lie Factor = 1.0

3. Form Follows Function

Design should support the purpose of the visualization.
First decide what the audience needs to do, then choose a design that helps them do it.
🧠 Function First, Form Second

4. Cole Nussbaumer's Four Design Concepts

Concept Purpose
Affordances Guide audience attention
Accessibility Easy for everyone to understand
Aesthetics Visually appealing
Acceptance Encourage stakeholder buy-in

5. Affordances

Design cues that help viewers understand where to focus.
Highlight important information and eliminate distractions.
🧠 Direct the Eye

6. Accessibility

Visualizations should be understandable by people with different skills and backgrounds.
If users cannot read it, it fails.

7. Aesthetics & Acceptance

Good design attracts attention and increases audience acceptance.
Attractive visualizations encourage engagement and trust.
🧠 Beautiful + Useful = Powerful

8. Shaffer's 4 C's of Visualization

C Meaning
Clear Easy to understand
Clean Well organized
Concise Brief but complete
Captivating Engaging and memorable
Exam Shortcut:

4C = Clear + Clean + Concise + Captivating

9. Visualization by Analytical Task

Task Common Charts
Amounts Bar Charts
Distribution Histogram, Density Plot
Proportions Pie Chart, Stacked Bar
Relationships Scatter Plot, Bubble Plot
Geospatial Data Maps, Choropleths
Uncertainty Error Bars, Confidence Bands

10. Bar Charts

Used to compare quantities across categories.
Good Practice:
  • Start bars at zero
  • Label axes clearly
  • Use meaningful colors
Bars must start at ZERO.

11. Grouped vs Stacked Bars

Chart Best Use
Grouped Bar Compare categories directly
Stacked Bar Show part-to-whole relationships
🧠 Grouped = Compare
Stacked = Composition

12. Waterfall Chart

Shows cumulative positive and negative changes over time.
Profit changes across several months.

13. Dot Plot & Heatmap

Chart Strength
Dot Plot Highlights small differences
Heatmap Shows patterns in large datasets

14. Histograms & Density Plots

Used to display data distributions.
Chart Purpose
Histogram Shows frequencies using bins
Density Plot Smooth version of a histogram
Histograms are best for continuous numerical data.

15. Boxplots

Summarize a distribution using five key values.
Excellent for comparing multiple distributions.

16. Pie Charts

Show part-to-whole relationships.
Use only when:
  • Data sums to 100%
  • General comparison is sufficient
Avoid:
  • 3D pie charts
  • Too many slices
  • Precise comparisons

17. Scatter & Bubble Plots

Scatterplots show relationships between two numerical variables.
Useful for identifying:
  • Trends
  • Outliers
  • Clusters
Correlation ≠ Causation

18. Line Charts

Used to display trends over time or ordered data.
Use dashed lines to indicate missing data.

19. Slope Graph

Compares values across two points in time.
Excellent for showing increases and decreases.

20. Geospatial Visualizations

Chart Type Purpose
Map Show geographic information
Choropleth Map Color regions by value
Cartogram Resize regions by quantity
Heatmap Display intensity geographically

21. Final Exam Summary

Most Important Points

  • Form Follows Function — choose design based on purpose.
  • Affordances — guide audience attention.
  • Accessibility — keep visualizations readable and simple.
  • Shaffer 4C's — Clear, Clean, Concise, Captivating.
  • Bar Charts — start at zero.
  • Pie Charts — only for part-to-whole relationships.
  • Scatterplots — show relationships.
  • Line Charts — show trends over time.
  • Boxplots — compare distributions.
  • Maps & Choropleths — visualize geographic data.