Categorical scatter with jitter in Matplotlib - Step-by-Step Execution

Concept Flow - Categorical scatter with jitter

Start with categorical data

↓

Assign numeric positions to categories

↓

Add small random noise (jitter) to positions

↓

Plot points at jittered positions

↓

Display scatter plot with spread points

We convert categories to numbers, add small random shifts to avoid overlap, then plot points spread out for clarity.

Execution Sample

Matplotlib

import matplotlib.pyplot as plt
import numpy as np

categories = ['A', 'B', 'A', 'C', 'B']
positions = [0, 1, 0, 2, 1]
jitter = np.random.uniform(-0.1, 0.1, size=len(positions))
plt.scatter(np.array(positions) + jitter, [1]*len(positions))
plt.xticks([0,1,2], ['A','B','C'])
plt.show()

This code plots categorical points with jitter to spread overlapping points horizontally.

Execution Table

Step	Variable	Value/Action	Result/Effect
1	categories	['A', 'B', 'A', 'C', 'B']	Original categorical data
2	positions	[0, 1, 0, 2, 1]	Numeric positions assigned to categories
3	jitter	random values between -0.1 and 0.1	Small noise added to positions
4	positions + jitter	e.g. [0.05, 0.92, -0.07, 2.03, 1.08]	Jittered positions for plotting
5	plt.scatter	plot points at jittered positions	Scatter plot with spread points
6	plt.xticks	set x-axis labels to ['A','B','C']	Categories shown on x-axis
7	plt.show	display plot	Visual output with jittered categorical scatter

💡 All points plotted with jitter to avoid overlap, plot displayed.

Variable Tracker

Variable	Start	After Step 2	After Step 3	After Step 4	Final
categories	None	['A', 'B', 'A', 'C', 'B']	Same	Same	Same
positions	None	[0, 1, 0, 2, 1]	Same	Same	Same
jitter	None	None	[-0.07, -0.08, 0.05, 0.03, 0.08]	Same	Same
positions + jitter	None	None	None	[-0.07, 0.92, -0.07, 2.03, 1.08]	Same

Key Moments - 3 Insights

Why do we add jitter to the numeric positions?

Adding jitter spreads points horizontally so they don't overlap exactly, making each point visible. See execution_table step 4 where jitter is added.

How do categories become numbers for plotting?

Each category is assigned a number (like 0 for 'A', 1 for 'B', 2 for 'C') so matplotlib can plot them on a numeric axis. See execution_table step 2.

Why do we set x-axis ticks after plotting?

Because positions are numbers, we set x-axis labels to show category names for clarity. See execution_table step 6.

Visual Quiz - 3 Questions

Look at the variable_tracker table, what is the value of 'positions + jitter' after step 4?

A [0, 1, 0, 2, 1]

B ['A', 'B', 'A', 'C', 'B']

C [-0.07, 0.92, 0.05, 2.03, 1.08]

D [-0.1, 0.1, -0.1, 0.1, 0.1]

💡 Hint

Check the 'positions + jitter' row in variable_tracker after step 4.

At which step in the execution_table do we assign numeric positions to categories?

A Step 2

B Step 4

C Step 1

D Step 6

💡 Hint

Look for the step where 'positions' variable is set.

If jitter range was increased to (-0.5, 0.5), how would the scatter plot points change?

A Points would overlap more

B Points would be more spread out horizontally

C Points would move vertically

D Points would disappear

💡 Hint

Jitter controls horizontal noise added to positions, see execution_table step 3.

Concept Snapshot

Categorical scatter with jitter:
- Map categories to numbers (e.g., A=0, B=1)
- Add small random noise (jitter) to numeric positions
- Plot points at jittered positions to avoid overlap
- Set x-axis ticks to category labels
- Result: clear scatter plot showing category distribution

Full Transcript

This visual execution shows how to plot categorical data points with jitter using matplotlib. First, categories are converted to numeric positions. Then, small random noise called jitter is added to these positions to spread points horizontally and avoid overlap. The jittered positions are plotted as a scatter plot. Finally, x-axis ticks are set to show category names for clarity. This method helps visualize overlapping categorical points clearly.