Contents

Chapter 1

Choosing the Right Chart

Decision framework for selecting the most effective visualization for your data and message.

Quick Decision Tree

What are you showing?
β”‚
β”œβ”€β”€ COMPARISON between groups
β”‚   β”œβ”€β”€ Few categories (2-5) β†’ Bar chart (vertical)
β”‚   β”œβ”€β”€ Many categories (6+) β†’ Bar chart (horizontal)
β”‚   β”œβ”€β”€ Over time β†’ Line chart (grouped)
β”‚   └── Part of whole β†’ Stacked bar or pie (only if 2-5 segments)
β”‚
β”œβ”€β”€ DISTRIBUTION of values
β”‚   β”œβ”€β”€ Single variable β†’ Histogram + KDE
β”‚   β”œβ”€β”€ Compare distributions β†’ Box plot or violin plot
β”‚   β”œβ”€β”€ Exact percentiles matter β†’ Box plot
β”‚   └── Shape matters β†’ Violin plot or ridgeline
β”‚
β”œβ”€β”€ RELATIONSHIP between variables
β”‚   β”œβ”€β”€ Two numeric β†’ Scatter plot + regression line
β”‚   β”œβ”€β”€ Numeric vs categorical β†’ Strip/swarm plot
β”‚   β”œβ”€β”€ Many numeric pairs β†’ Pair plot or correlation heatmap
β”‚   └── Three variables β†’ Scatter with color/size encoding
β”‚
β”œβ”€β”€ TREND over time
β”‚   β”œβ”€β”€ One metric β†’ Line chart
β”‚   β”œβ”€β”€ Multiple metrics β†’ Multi-line or small multiples
β”‚   β”œβ”€β”€ Cumulative β†’ Area chart
β”‚   └── Seasonal patterns β†’ Seasonal decomposition plot
β”‚
└── COMPOSITION
    β”œβ”€β”€ At one point in time β†’ Pie chart (≀5 segments) or treemap
    β”œβ”€β”€ Over time β†’ Stacked area
    └── Hierarchical β†’ Treemap or sunburst

Rules of Thumb

SituationBest ChartAvoid
Comparing exact valuesBar chartPie chart
Showing trendLine chartBar chart
Showing distributionHistogramBar chart
CorrelationsScatter plotTable
Part-of-whole (≀5 parts)Pie/donutBar chart (sometimes)
Part-of-whole (>5 parts)TreemapPie chart
Before/after comparisonPaired bar or slope chartGrouped bars
Geographic dataChoropleth mapBar chart by region

Common Anti-Patterns

1. Dual Y-axes β€” Almost always misleading. Use small multiples instead.

2. 3D charts β€” Never add depth for decoration. It distorts perception.

3. Pie charts with >5 segments β€” Humans can't accurately compare angles.

4. Truncated Y-axis without indication β€” Either start at 0 or clearly mark the break.

5. Rainbow color schemes — Use sequential (light→dark) for ordered data, categorical for groups.

Color Guidelines

  • Sequential data (lowβ†’high): Use single-hue gradient (light blue β†’ dark blue)
  • Diverging data (negative↔positive): Use two-hue gradient (red ← white β†’ blue)
  • Categorical data: Use distinct hues with similar saturation
  • Highlighting: Use gray for context, color for focal point
  • Accessibility: Always check with colorblind simulation (use viridis for safe default)
Chapter 2
πŸ”’ Available in full product

Statistical Tests Cheatsheet

Chapter 3
πŸ”’ Available in full product

Choosing the Right Chart

Chapter 4
πŸ”’ Available in full product

Statistical Tests Cheatsheet

You’ve reached the end of the free preview

Get the full Python Data Analysis Toolkit and unlock everything.

All Chapters

Get the complete guide with every chapter unlocked, including code samples, diagrams, and best practices.

Full Tool Suite

Access all interactive tools with complete data, all workload profiles, and the full scenario library.

Source Files

Downloadable source code, configuration files, and working examples from every chapter.

Lifetime Updates

Free updates for life. Every new chapter, tool, and improvement included.

Buy Now — $39 →
πŸ“¦ Free sample included — download another copy or visit the store for the full product.
Python Data Analysis Toolkit v1.0.0 β€” Free Preview