color-accessibility-guide

Color Accessibility Guide

Design data visualizations that are accessible to colorblind readers and follow best practices for clarity, using tested palettes and encoding principles.

Color Vision Deficiency Overview

Approximately 8% of males and 0.5% of females have some form of color vision deficiency (CVD). The most common types:

Type	Prevalence (Male)	Affected Colors	Commonly Confused
Deuteranomaly (green-weak)	5%	Green	Red and green
Protanomaly (red-weak)	1%	Red	Red and green
Deuteranopia (no green)	1%	Green	Red and green
Protanopia (no red)	1%	Red	Red and green
Tritanopia (no blue)	0.003%	Blue	Blue and yellow
Monochromacy	Very rare	All	All colors

Key takeaway: Never rely solely on a red-green distinction to convey information. About 1 in 12 male readers cannot distinguish them.

Recommended Colorblind-Safe Palettes

Qualitative Palettes (Categorical Data)

Wong (2011) Nature Palette (8 colors)

Widely recommended for scientific publications:

# Wong's colorblind-friendly palette
wong_palette = {
    "black":       "#000000",
    "orange":      "#E69F00",
    "sky_blue":    "#56B4E9",
    "bluish_green":"#009E73",
    "yellow":      "#F0E442",
    "blue":        "#0072B2",
    "vermillion":  "#D55E00",
    "reddish_purple":"#CC79A7"
}

Okabe-Ito Palette

okabe_ito = ["#E69F00", "#56B4E9", "#009E73", "#F0E442",
             "#0072B2", "#D55E00", "#CC79A7", "#000000"]

Tol's Qualitative Palette

# Paul Tol's qualitative palette (up to 12 distinct colors)
tol_qualitative = ["#332288", "#88CCEE", "#44AA99", "#117733",
                   "#999933", "#DDCC77", "#CC6677", "#882255",
                   "#AA4499", "#661100", "#6699CC", "#888888"]

Sequential Palettes (Ordered Data)

For continuous data, use perceptually uniform colormaps:

import matplotlib.pyplot as plt

# Recommended sequential colormaps
# These are perceptually uniform and colorblind-safe:
good_cmaps = ["viridis", "plasma", "inferno", "magma", "cividis"]

# Avoid these (not perceptually uniform, not colorblind-safe):
bad_cmaps = ["jet", "rainbow", "hsv"]  # NEVER use these

# Example usage
import numpy as np
data = np.random.randn(10, 10)
fig, ax = plt.subplots(figsize=(8, 6))
im = ax.imshow(data, cmap="viridis")
plt.colorbar(im)
plt.title("Use viridis, not jet")
plt.savefig("heatmap.pdf", dpi=300, bbox_inches="tight")