[raincloud-basic] Raincloud Plot - Combined violin, jitter points, and box plot

[MarkusNeusinger]

A raincloud plot combining half-violin/KDE ('cloud'), jittered strip plot ('rain'), and box plot to show distribution shape alongside raw data points.

Why important

• Called the 'gold standard' for statistical visualization
• Solves the problem that boxplots hide bimodality
• Shows both summary statistics AND raw data

Library

• Primary: ptitprince (wrapper for seaborn/matplotlib)

Applications

• Scientific publications requiring transparent statistical visualization
• Comparing distributions across groups
• Revealing hidden patterns in data distribution

[claude]

✅ Spec ID: raincloud-basic

Rationale: Raincloud plots are a well-established visualization combining half-violin, jittered strip points, and box plot—distinct from our existing violin-basic, strip-basic, and box-basic specs which each show only one component.

Add label approved to begin generation.

[github-actions]

✅ Specification Ready: raincloud-basic

PR: #1881

---

specification.md

raincloud-basic: Basic Raincloud Plot

Description

A raincloud plot combines three visualization elements—a half-violin (the "cloud"), jittered strip points (the "rain"), and a box plot—to provide a comprehensive view of data distribution. This hybrid approach shows distribution shape, summary statistics, and individual observations simultaneously, addressing the limitation of box plots that can hide multimodal distributions. Often called the "gold standard" for transparent statistical visualization in scientific publications.

Applications

• Comparing treatment effects across experimental groups in clinical trials
• Visualizing response time distributions in psychology experiments
• Analyzing score distributions across different conditions in A/B testing
• Presenting survey response distributions by demographic groups

Data

• category (categorical) - Group labels for comparison on the categorical axis
• value (numeric) - Continuous variable values shown on the value axis
• Size: 20-300 observations per category, 2-6 categories
• Example: Reaction times (ms) for control vs treatment groups

Notes

• Position half-violin on one side, jittered points on the other, with box plot in between
• Use moderate jitter (0.05-0.1) to spread points without excessive overlap
• Apply transparency (alpha 0.5-0.7) to jittered points
• Include median and quartile markers in box plot
• Consider horizontal orientation for better label readability with many categories

---

Next: Add approved label to merge this specification to main.

---

🤖 spec-create workflow

[github-actions]

🚀 Specification Merged!

Specification: raincloud-basic
PR: #1881 (merged)

The specification is now in main and synced to PostgreSQL.

---

Next: Add generate:{library} labels to start generating implementations.

Available libraries: matplotlib, seaborn, plotly, bokeh, altair, plotnine, pygal, highcharts, letsplot

Or use workflow dispatch for bulk operations.

---

🤖 spec-create workflow

[MarkusNeusinger]

✅ Closed with 10/9 implementations in repo (raincloud-basic)