Categorical plots (boxplot, violinplot) in Data Analysis Python

Introduction

Categorical plots help us see how data values spread and differ across groups. Boxplots and violinplots show this clearly for categories.

Comparing test scores of students from different classes.

Checking how sales vary by product category.

Seeing the distribution of heights for men and women.

Understanding customer ratings across different stores.

Syntax

Data Analysis Python

import seaborn as sns
sns.boxplot(x='category_column', y='value_column', data=dataframe)
sns.violinplot(x='category_column', y='value_column', data=dataframe)

Use boxplot to see median, quartiles, and outliers.

Use violinplot to see data distribution shape and density.

Examples

Shows boxplot of sepal length for each iris species.

Data Analysis Python

sns.boxplot(x='species', y='sepal_length', data=iris_df)

Shows violinplot of total bill amounts for each day of the week.

Data Analysis Python

sns.violinplot(x='day', y='total_bill', data=tips_df)

Compares height distributions between genders using boxplots.

Data Analysis Python

sns.boxplot(x='gender', y='height', data=people_df)

Sample Program

This code loads penguin data and shows two plots: a boxplot and a violinplot of flipper length for each species. It helps compare the spread and distribution of flipper lengths.

Data Analysis Python

import seaborn as sns
import matplotlib.pyplot as plt

# Load example dataset
penguins = sns.load_dataset('penguins')

# Create boxplot
plt.figure(figsize=(8,4))
sns.boxplot(x='species', y='flipper_length_mm', data=penguins)
plt.title('Boxplot of Flipper Length by Penguin Species')
plt.show()

# Create violinplot
plt.figure(figsize=(8,4))
sns.violinplot(x='species', y='flipper_length_mm', data=penguins)
plt.title('Violinplot of Flipper Length by Penguin Species')
plt.show()

Important Notes

Boxplots summarize data with median and quartiles, good for spotting outliers.

Violinplots add detail by showing data distribution shape, useful for seeing if data is skewed or has multiple peaks.

Both plots require categorical data on one axis and numeric data on the other.

Summary

Categorical plots help compare groups easily.

Boxplots show summary statistics and outliers.

Violinplots show data distribution shapes.

Practice

1. What does a boxplot primarily show about a dataset?

easy

A. The correlation between two variables

B. The median, quartiles, and outliers of the data

C. Only the mean and standard deviation

D. The exact values of all data points

Solution

1. Step 1: Understand boxplot components

   A boxplot displays the median line, the box showing the interquartile range (25th to 75th percentile), and whiskers for range excluding outliers.

2. Step 2: Identify what it does not show

   It does not show every data point's exact value or correlation between variables.

3. Final Answer:

   The median, quartiles, and outliers of the data -> Option B

4. Quick Check:

   Boxplot = median + quartiles + outliers [OK]

Hint: Boxplots summarize data spread with median and quartiles [OK]

Common Mistakes:

• Thinking boxplots show all individual data points
• Confusing boxplots with scatterplots
• Assuming boxplots show mean instead of median

2. Which of the following is the correct way to create a violin plot using seaborn for a DataFrame df with columns 'category' and 'value'?

easy

A. sns.violinplot(y='category', x='value', data=df)

B. sns.violinplot(data=df, x='value', y='category')

C. sns.violinplot(df['category'], df['value'])

D. sns.violinplot(x='category', y='value', data=df)

Solution

1. Step 1: Check seaborn violinplot syntax

   The correct syntax is sns.violinplot(x='category', y='value', data=df) where x is the categorical variable and y is the numeric variable.

2. Step 2: Identify incorrect options

   sns.violinplot(data=df, x='value', y='category') swaps x and y incorrectly. sns.violinplot(df['category'], df['value']) passes series without specifying data keyword. sns.violinplot(y='category', x='value', data=df) swaps x and y axes.

3. Final Answer:

   sns.violinplot(x='category', y='value', data=df) -> Option D

4. Quick Check:

   Violinplot syntax = sns.violinplot(x=cat, y=num, data=df) [OK]

Hint: Use x=category, y=value, data=DataFrame for seaborn violinplot [OK]

Common Mistakes:

• Swapping x and y parameters
• Passing columns without data= argument
• Using positional arguments instead of keywords

3. Given the following code, what will be the output type?

import seaborn as sns
import pandas as pd
import matplotlib.pyplot as plt

# Sample data
df = pd.DataFrame({
    'group': ['A', 'A', 'B', 'B', 'C', 'C'],
    'score': [10, 15, 10, 20, 15, 25]
})

plot = sns.boxplot(x='group', y='score', data=df)
plt.show()

medium

A. A boxplot showing score distribution per group

B. A scatter plot of scores vs groups

C. A line plot connecting scores in each group

D. An error because of missing hue parameter

Solution

1. Step 1: Understand sns.boxplot usage

   The code uses seaborn's boxplot with x as categorical 'group' and y as numeric 'score', which creates boxplots per group.

2. Step 2: Confirm plot type and parameters

   No hue is required here; the plot will show boxplots for each group category with score distribution.

3. Final Answer:

   A boxplot showing score distribution per group -> Option A

4. Quick Check:

   sns.boxplot(x=cat, y=num) = boxplot per category [OK]

Hint: sns.boxplot(x=cat, y=num) plots distribution per category [OK]

Common Mistakes:

• Expecting scatter or line plot instead of boxplot
• Thinking hue is mandatory
• Confusing boxplot with barplot

4. Identify the error in this code snippet for creating a violin plot:

import seaborn as sns
sns.violinplot(data=df, x='value', y='category')

medium

A. DataFrame df is not defined

B. Missing import for matplotlib.pyplot

C. The x and y parameters are swapped; x should be categorical

D. Violinplot does not accept x and y parameters

Solution

1. Step 1: Check parameter roles in violinplot

   In seaborn violinplot, x should be the categorical variable and y the numeric variable. Here, 'value' is numeric and 'category' is categorical, so x and y are swapped.

2. Step 2: Verify other possible errors

   Importing matplotlib is not mandatory for just creating the plot object. The DataFrame df is assumed defined. Violinplot accepts x and y parameters.

3. Final Answer:

   The x and y parameters are swapped; x should be categorical -> Option C

4. Quick Check:

   Violinplot x=category, y=numeric [OK]

Hint: Categorical variable must be x in violinplot [OK]

Common Mistakes:

• Swapping x and y axes
• Assuming matplotlib import is always needed
• Not defining DataFrame before plotting

5. You have a dataset with three groups: 'X', 'Y', and 'Z'. Group 'Z' has many outliers. You want to visualize the distribution shape and highlight the outliers clearly. Which plot should you choose and why?

hard

A. Use a violinplot to show distribution shape and a boxplot overlay to highlight outliers

B. Use only a boxplot because violinplots hide outliers

C. Use a scatter plot to show all points without summary

D. Use a line plot to connect group means

Solution

1. Step 1: Understand plot strengths

   Violinplots show the shape and density of data distribution well, but outliers can be less obvious. Boxplots clearly mark outliers.

2. Step 2: Combine plots for best insight

   Overlaying a boxplot on a violinplot combines both benefits: distribution shape and clear outlier marking.

3. Final Answer:

   Use a violinplot to show distribution shape and a boxplot overlay to highlight outliers -> Option A

4. Quick Check:

   Violin + boxplot = shape + outliers [OK]

Hint: Combine violin and boxplot for shape and outliers [OK]

Common Mistakes:

• Thinking violinplots alone highlight outliers well
• Using scatter plots without summary
• Choosing line plots for categorical distributions