Categorical Plots: boxplot, violinplot, and countplot

Reading Time: 8 Mins | Level: Intermediate

1. Comparing Distributions Across Categorical Groups

To compare numerical distributions across discrete categories, Seaborn provides:

• sns.boxplot(): Shows the 5-number statistical summary (Min, Q1, Median, Q3, Max) and identifies outliers.
• sns.violinplot(): Combines a box plot with a rotated kernel density curve to reveal multimodal shapes.
• sns.countplot(): Displays frequency counts for categorical variables (like an automated value_counts()).

2. Anatomy of a Box Plot

A Box Plot (Box-and-Whisker) provides 5 key metrics:

1. 1
   Median (Q2 / 50th percentile): The central line inside the box.
2. 2
   First Quartile (Q1 / 25th percentile): Bottom of the box.
3. 3
   Third Quartile (Q3 / 75th percentile): Top of the box.
4. 4
   Interquartile Range (IQR): $IQR = Q3 - Q1$ (the middle 50% of the data).
5. 5
   Whiskers: Extend to $1.5 imes IQR$ from Q1 and Q3. Points beyond whiskers are plotted as individual outlier markers!