boxplot() Method¶

The boxplot() method creates box-and-whisker plots that summarize the distribution of numeric data, showing median, quartiles, and outliers.

Anatomy of a Box Plot¶

    ┌─────────┐
    │         │   ← Maximum (or upper fence)
    │    ○    │   ← Outliers (beyond 1.5×IQR)
    │    │    │
────┼────┼────┼── ← Q3 (75th percentile)
    │    │    │
    │    │    │   ← IQR (Interquartile Range)
    │    │    │
────┼────┼────┼── ← Median (Q2, 50th percentile)
    │    │    │
    │    │    │
────┼────┼────┼── ← Q1 (25th percentile)
    │    │    │
    │    │    │
    └────┴────┘   ← Minimum (or lower fence)

Basic Usage¶

Single Column¶

import pandas as pd
import matplotlib.pyplot as plt

url = "https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv"
df = pd.read_csv(url)

fig, ax = plt.subplots(figsize=(5, 4))
df.boxplot(column='Age', ax=ax)
ax.set_title('Age Distribution')
plt.show()

Multiple Columns¶

fig, ax = plt.subplots(figsize=(10, 5))
df.boxplot(column=['Age', 'Fare'], ax=ax)
ax.set_title('Age and Fare Distributions')
plt.show()

Key Parameters¶

column - Select Columns¶

# Single column
df.boxplot(column='Age')

# Multiple columns
df.boxplot(column=['Age', 'Fare', 'SibSp'])

by - Group by Category¶

Create separate box plots for each category:

fig, ax = plt.subplots(figsize=(10, 5))
df.boxplot(column='Age', by='Pclass', ax=ax)
ax.set_title('Age Distribution by Passenger Class')
plt.suptitle('')  # Remove automatic title
plt.show()

vert - Orientation¶

# Vertical (default)
df.boxplot(column='Age', vert=True)

# Horizontal
fig, ax = plt.subplots(figsize=(8, 3))
df['Age'].plot(kind='box', vert=False, ax=ax)
ax.set_title('Horizontal Box Plot')
plt.show()

figsize - Figure Size¶

df.boxplot(column='Age', figsize=(8, 5))

Practical Example: Titanic Passenger Ages¶

url = "https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv"
df = pd.read_csv(url, index_col='PassengerId')

fig, ax = plt.subplots(figsize=(8, 4))
df['Age'].plot(kind='box', ax=ax, vert=False)
ax.set_title('Horizontal Boxplot of Passenger Ages on Titanic')
ax.set_xlabel('Age')
ax.spines[['top', 'left', 'right']].set_visible(False)
plt.show()

Grouped Box Plots¶

By Single Category¶

fig, ax = plt.subplots(figsize=(10, 5))
df.boxplot(column='Fare', by='Pclass', ax=ax)
ax.set_ylabel('Fare')
plt.suptitle('')
ax.set_title('Fare Distribution by Class')
plt.show()

By Multiple Categories¶

fig, ax = plt.subplots(figsize=(12, 5))
df.boxplot(column='Age', by=['Pclass', 'Survived'], ax=ax)
plt.suptitle('')
ax.set_title('Age by Class and Survival')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

Comparing Multiple Variables¶

url = 'https://raw.githubusercontent.com/justmarkham/DAT8/master/data/drinks.csv'
df = pd.read_csv(url)

fig, ax = plt.subplots(figsize=(10, 5))
df[['beer_servings', 'spirit_servings', 'wine_servings']].boxplot(ax=ax)
ax.set_title('Alcohol Servings by Type')
ax.set_ylabel('Servings')
plt.show()

Customization¶

Using Return Value¶

fig, ax = plt.subplots(figsize=(8, 5))
bp = df.boxplot(column='Age', ax=ax, return_type='dict')

# Customize colors
for box in bp['boxes']:
    box.set_color('steelblue')
for whisker in bp['whiskers']:
    whisker.set_color('gray')
for cap in bp['caps']:
    cap.set_color('gray')
for median in bp['medians']:
    median.set_color('red')
    median.set_linewidth(2)

plt.show()

Grid and Appearance¶

fig, ax = plt.subplots(figsize=(8, 5))
df.boxplot(
    column='Age',
    ax=ax,
    grid=False,
    notch=True,  # Confidence interval notch
    patch_artist=True  # Enable fill color
)
plt.show()

boxplot() vs plot(kind='box')¶

Both methods create box plots, but with slightly different interfaces:

# Using boxplot()
df.boxplot(column='Age')
df.boxplot(column='Age', by='Pclass')

# Using plot(kind='box')
df['Age'].plot(kind='box')
df[['Age', 'Fare']].plot(kind='box')

Feature	boxplot()	plot(kind='box')
by parameter	✅ Direct support	❌ Manual grouping
Multiple columns	column=list	Select columns first
return_type	✅ Supported	❌ Not supported
Method location	DataFrame only	Series and DataFrame

Interpreting Box Plots¶

# Get the statistics shown in box plot
stats = df['Age'].describe()
print(stats)

count    714.000000
mean      29.699118
std       14.526497
min        0.420000
25%       20.125000    ← Q1 (box bottom)
50%       28.000000    ← Median (line in box)
75%       38.000000    ← Q3 (box top)
max       80.000000

Identifying Outliers¶

Outliers are points beyond: - Upper fence: Q3 + 1.5 × IQR - Lower fence: Q1 - 1.5 × IQR

Q1 = df['Age'].quantile(0.25)
Q3 = df['Age'].quantile(0.75)
IQR = Q3 - Q1

lower_fence = Q1 - 1.5 * IQR
upper_fence = Q3 + 1.5 * IQR

outliers = df[(df['Age'] < lower_fence) | (df['Age'] > upper_fence)]
print(f"Number of outliers: {len(outliers)}")

Method Signature¶

DataFrame.boxplot(
    column=None,       # Column(s) to plot
    by=None,           # Group by column
    ax=None,           # Matplotlib axes
    fontsize=None,     # Tick label font size
    rot=0,             # Tick label rotation
    grid=True,         # Show grid
    figsize=None,      # Figure size
    layout=None,       # Subplot layout
    return_type=None,  # Return type ('axes', 'dict', 'both')
    **kwargs           # Additional kwargs
)

Summary¶

# Single column
df.boxplot(column='Age')

# Multiple columns
df.boxplot(column=['Age', 'Fare'])

# Grouped by category
df.boxplot(column='Age', by='Pclass')

# Horizontal (via plot)
df['Age'].plot(kind='box', vert=False)

# Customized
df.boxplot(column='Age', grid=False, notch=True)