Audit 'ggplot2' Visualizations for Accessibility and Best Practices.

GGenemy

Enemy of Bad Practices in Data Visualization

GGenemy is an R package that audits your ggplot2 visualizations for accessibility issues, misleading practices, and readability problems. Think of it as a linter for your data visualizations.

Why GGenemy?

Data visualizations can inadvertently mislead or exclude audiences through:

• Poor color choices that are invisible to colorblind users (~8% of males)
• Misleading scales like truncated axes or dual y-axes
• Readability issues like tiny text or overlapping labels
• Accessibility barriers that make charts hard to interpret

GGenemy catches these issues automatically and suggests fixes.

Installation

# Install from CRAN (coming soon)
install.packages("GGenemy")

# Or install development version from GitHub:
# install.packages("devtools")
devtools::install_github("andytai7/GGenemy")

Quick Start

library(GGenemy)
library(ggplot2)

# Create a plot with some issues
p <- ggplot(mtcars, aes(wt, mpg, color = factor(cyl))) +
  geom_point() +
  scale_color_manual(values = c("red", "green", "blue"))

# Run comprehensive audit
report <- gg_audit(p)
print(report)

Key Features

Comprehensive Auditing

# Check everything
gg_audit(your_plot)

# Check specific aspects
gg_audit(your_plot, checks = c("color", "scales"))

Color Accessibility

Detects problematic color combinations and suggests colorblind-safe alternatives:

gg_audit_color(your_plot)

Colorblind Simulation

See how your plot appears to colorblind users:

# Simulate different types of color vision deficiency
gg_simulate_cvd(your_plot, type = "deutan")  # green-blind
gg_simulate_cvd(your_plot, type = "protan")  # red-blind
gg_simulate_cvd(your_plot, type = "tritan")  # blue-blind

Scale & Axis Checks

Catches misleading practices:

• Truncated axes on bar charts
• Dual y-axes
• Inappropriate aspect ratios
• Unlabeled log scales

gg_audit_scales(your_plot)

Automatic Fixes

Get code suggestions or apply fixes automatically:

# Get copy-paste code suggestions
gg_suggest_fixes(your_plot)

# Apply automatic fixes
fixed_plot <- gg_suggest_fixes(your_plot, auto_fix = TRUE)

Accessibility Auditing

Checks for:

• Text size and readability
• Point/line sizes
• Redundant encoding (not relying on color alone)
• Label clarity

gg_audit_accessibility(your_plot)

Example Workflow

library(ggplot2)
library(GGenemy)

# Create a problematic plot
problematic_plot <- ggplot(mtcars, aes(x = wt, y = mpg, color = factor(cyl))) +
  geom_point(size = 1) +
  scale_color_manual(values = c("red", "green", "blue")) +
  labs(title = "Plot", x = "wt", y = "mpg")

# Audit it
report <- gg_audit(problematic_plot)

# Get fix suggestions
gg_suggest_fixes(problematic_plot)

# Apply automatic fixes
improved_plot <- gg_suggest_fixes(problematic_plot, auto_fix = TRUE)

# View the improved plot
print(improved_plot)

Philosophy

GGenemy believes in:

1. Accessibility First: Visualizations should be perceivable by everyone
2. Honest Representation: Data should be presented without distortion
3. Clarity Over Aesthetics: Readability trumps flashiness
4. Constructive Guidance: We suggest fixes, not just criticisms

Roadmap

• [x] Color accessibility checking
• [x] Colorblind vision simulation
• [x] Scale and axis auditing
• [x] Text readability checks
• [x] Automatic fix suggestions
• [ ] Integration with RMarkdown/Quarto for batch auditing
• [ ] Custom rule sets
• [ ] Interactive Shiny app for exploration

Contributing

GGenemy is in active development! Contributions are welcome:

• Report bugs
• Suggest features
• Improve documentation
• Submit pull requests

Related Packages

• colorspace - Color vision deficiency simulation
• viridis - Colorblind-safe palettes
• ggplot2 - The grammar of graphics

Citation

citation("GGenemy")

License

MIT © Andy Man Yeung Tai

GGenemy: Making data visualization more accessible, honest, and clear.