Distribution
Split violin plot
Compares two subgroup distributions using opposite halves of a violin within each category.
- Plot
- Distribution
- Comparison
What it answers
How do two subgroups differ within each category?
- Shape
- Opposing half-violins
- What it shows
- Within-category shifts and shape differences.
Example
The question
Did handling times change after the new process, in each team?
Minutes to handle a case, before and after the change (split violins, illustrative)
Illustrative data. It belongs to no company.
When to use it
Use it when
Two subgroup distributions per category.
Avoid it when
More than two subgroups or unclear left/right mapping.
How to read it
Compare left and right halves at the same value.
Note
Normally exactly two subgroups per category.
The data you need
- Layout
- Long table
- Fields
- Category: text; Subgroup: text; Value: number
- One row is
- One raw observation for a category and subgroup. What is the grain?
- Rules
- Use two subgroups per category for a conventional split violin. Keep raw numeric observations.
- Optional
- For box plots only, some tools accept Group,Q1,Median,Q3,WhiskerLow,WhiskerHigh plus separate outliers. Other distributions usually need raw data.
- Preparation
- Keep raw numeric observations, group them if needed and handle missingness explicitly. Choose bins, bandwidth or summary conventions as appropriate to this specific plot.
- Units
- The measurement axis retains its original unit. Frequency uses observations; normalised frequency divides by group sample size; density also depends on interval width.
Worked example
CategoryA has Before values 10,12 and After 18,23. Put halves side by side; the observed After values are higher. More data is needed for credible density estimation.
| Category | Subgroup | Value |
|---|---|---|
| A | Before | 10 |
| A | Before | 12 |
| A | After | 18 |
| A | After | 23 |