Scatter Plots

Learn how to use scatter plots to study relationships between two numerical variables.

Skill Level: Beginner
Prerequisites: Line Charts
Estimated Time: 25 minutes

Story Time

So far, you have learned how to compare categories with bar charts and how to show change over time with line charts.

Now we move to another important analytical question: Are two numerical variables related?

For example:

• Do students with higher reading scores also tend to have higher writing scores?
• Is there a relationship between hours studied and exam performance?
• Do two measurements increase together, decrease together, or show no clear pattern?

Scatter plots help answer these questions.

In this lesson, you will learn how to create scatter plots with Matplotlib and how to interpret the patterns they show.

Learning Objectives

By the end of this lesson, you will be able to:

• explain when a scatter plot should be used
• create a scatter plot with Matplotlib
• add titles and axis labels
• interpret positive, negative, and weak relationships
• recognize outliers and clusters
• avoid common mistakes when reading scatter plots

Why This Topic Matters

Many real data problems involve relationships between two numerical variables.

Examples include:

• study time and exam score
• reading score and writing score
• advertising spend and sales
• temperature and electricity usage

A scatter plot makes these relationships easier to see by plotting each observation as a point. It helps analysts understand whether two variables tend to move together and whether the relationship appears strong, weak, or unclear.

What Is a Scatter Plot?

A scatter plot shows pairs of numerical values on an x-y coordinate system.

Each point represents one observation:

• the x-position shows the value of one variable
• the y-position shows the value of the second variable

For example, one point might represent a student with:

• reading score = 72
• writing score = 75

So every dot on the chart corresponds to one row of data.

When Should You Use a Scatter Plot?

Use a scatter plot when:

• both variables are numerical
• you want to study whether they are related
• you want to look for patterns, clusters, or outliers

A scatter plot is especially useful when the question is about relationship, not comparison or trend.

For example:

• reading score vs writing score → scatter plot
• gender vs average math score → bar chart
• sales over months → line chart

Your First Scatter Plot

You can create a scatter plot in Matplotlib with plt.scatter().

import matplotlib.pyplot as plt

reading = 
writing = 

plt.scatter(reading, writing)
plt.show()

This creates a scatter plot where:

• the x-axis shows reading scores
• the y-axis shows writing scores

Each point represents one student or one observation.

Adding a Title and Axis Labels

As with other charts, labels make the plot easier to interpret.

import matplotlib.pyplot as plt

reading = 
writing = 

plt.scatter(reading, writing)
plt.title("Reading Score vs Writing Score")
plt.xlabel("Reading Score")
plt.ylabel("Writing Score")
plt.show()

Titles and axis labels help viewers understand what each axis represents, which is essential for correct interpretation.

Loading interactive scatter plot…

Interpreting the Direction of a Relationship

One of the first things to notice in a scatter plot is the direction of the relationship.

Positive Relationship

If points tend to move upward from left to right, the relationship is positive.

This means:

• as x increases
• y also tends to increase

For example, if students with higher reading scores also tend to have higher writing scores, the scatter plot will show an upward pattern.

Negative Relationship

If points tend to move downward from left to right, the relationship is negative.

This means:

• as x increases
• y tends to decrease

An example might be a case where more absences are associated with lower scores.

No Clear Relationship

If the points appear scattered without a clear upward or downward pattern, there may be little or no relationship between the variables.

This means knowing one variable does not strongly help you predict the other.

Interpreting Strength

The strength of the relationship depends on how closely the points follow a pattern.

• If the points are tightly grouped around an upward or downward path, the relationship is stronger.
• If the points are spread widely, the relationship is weaker.

When reading a scatter plot, try not to focus on one point alone. Look at the overall arrangement of the points.

Outliers and Unusual Points

A scatter plot can also reveal outliers.

An outlier is a point that lies far from the main group of points.

For example, if most students with reading scores around 80 also have writing scores around 78–85, but one student has a writing score of 40, that point may be unusual.

Outliers are important because they may represent:

• data entry errors
• unusual observations
• or important exceptions in the data

Example: Positive Relationship

Suppose you create a scatter plot of reading and writing scores.

import matplotlib.pyplot as plt

reading = 
writing = 

plt.scatter(reading, writing)
plt.title("Reading Score vs Writing Score")
plt.xlabel("Reading Score")
plt.ylabel("Writing Score")
plt.show()

A good interpretation would be:

• the points show an upward pattern
• students with higher reading scores also tend to have higher writing scores
• the relationship appears positive

Example: Weak Relationship

Now imagine a different scatter plot:

import matplotlib.pyplot as plt

hours_studied =[1][2][3][4][5][6]
quiz_score =[7]

plt.scatter(hours_studied, quiz_score)
plt.title("Hours Studied vs Quiz Score")
plt.xlabel("Hours Studied")
plt.ylabel("Quiz Score")
plt.show()

This chart may still suggest some upward tendency, but the points are more spread out.

A stronger interpretation would be:

• there may be a positive relationship
• but it does not appear very strong
• the scores vary even for similar study hours

Customizing a Scatter Plot

You can adjust the color, size, and transparency of points.

import matplotlib.pyplot as plt

reading = 
writing = 

plt.scatter(reading, writing, color="purple", s=80, alpha=0.7)
plt.title("Reading Score vs Writing Score")
plt.xlabel("Reading Score")
plt.ylabel("Writing Score")
plt.show()

In this example:

• color="purple" changes the point color
• s=80 changes the point size
• alpha=0.7 makes points slightly transparent

These choices can improve readability, especially when many points overlap.

Reading a Scatter Plot as an Analyst

When interpreting a scatter plot, ask:

1. What variable is on the x-axis?
2. What variable is on the y-axis?
3. Is the relationship positive, negative, or unclear?
4. Is the relationship strong or weak?
5. Are there clusters of points?
6. Are there outliers or unusual observations?
7. What is the main message in one sentence?

For example:

Reading and writing scores show a positive relationship, with higher reading scores generally associated with higher writing scores.

That is a more analytical interpretation than simply saying, “The points go upward.”

Google Colab Activity

Using the Students Performance dataset, create a scatter plot of:

• reading score on the x-axis
• writing score on the y-axis

Then answer:

• Does the relationship appear positive, negative, or unclear?
• Does it seem strong or weak?
• Are there any unusual points?

Practice Exercise

Create a scatter plot using two numerical variables from a dataset you are working with.

Then write 2–3 sentences that describe:

• the direction of the relationship
• the strength of the relationship
• whether any outliers appear

Practice in Colab

Notebook

Open in Google Colab

Practice scatter plots with short runnable examples and exercises.

After completing the activity in Google Colab, mark it as complete below.

Self Evaluation

Check Your Understanding

1. When should you use a scatter plot?

When comparing categoriesWhen showing change over time onlyWhen studying the relationship between two numerical variablesWhen counting how many items belong to each group

2. Which Matplotlib function creates a scatter plot?

plt.plot()plt.bar()plt.scatter()plt.hist()

3. What does a positive relationship look like in a scatter plot?

Points trend upward from left to rightPoints trend downward from left to rightPoints form separate barsPoints are arranged in bins

4. What is an outlier in a scatter plot?

A point that lies far from the main cluster of pointsA point that is exactly in the centerA category with the highest countA missing axis label

5. Which statement is correct about scatter plots?

They prove causation between variablesThey show relationships, but not necessarily causationThey should only be used for categorical dataThey are best for showing frequency counts

Challenge Exercise

Create a scatter plot that tells a meaningful story about student performance.

Your chart should include:

• clear axis labels
• a title
• readable plotting choices
• a short written interpretation

Try to explain not just what the chart is, but what it suggests about the relationship between the two variables.

Common Mistakes

Using a Scatter Plot with Categorical Variables

Scatter plots are meant for two numerical variables. If one variable is categorical, another chart type is usually better.

Looking at Individual Points Only

The value of a scatter plot comes from the overall pattern, not just one or two dots. Look for direction, strength, and clustering.

Assuming Correlation Means Causation

A scatter plot may show that two variables are related, but it does not prove that one causes the other.

Ignoring Outliers

Unusual points can be important. They may reflect errors, rare cases, or important exceptions that deserve attention.

Key Takeaways

In this lesson, you learned how to:

• create scatter plots with Matplotlib
• use scatter plots to study relationships between two numerical variables
• interpret positive, negative, and weak relationships
• identify outliers and clusters
• read scatter plots more carefully as an analyst

Continue Your Journey

Next, you will learn Histograms, which help you understand how a single numerical variable is distributed.