Introduction to Data Visualisation

Learn what data visualisation is, why it matters, and how charts help you answer different types of analytical questions.

Skill Level: Beginner
Prerequisites: Working with Data module
Estimated Time: 20 minutes

Story Time

You have already learned how to load, clean, and summarize data. But numbers in a table do not always make patterns easy to see. If you want to understand trends, compare groups, or communicate findings clearly, you need something more visual.

In this lesson, you will learn what data visualisation is, why it matters, what kinds of questions charts can answer, and how to begin reading charts as an analyst.

Learning Objectives

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

• explain what data visualisation is
• describe why visualisation is important in data analysis
• identify common analytical questions that charts help answer
• distinguish between comparison, trend, distribution, and relationship
• begin reading and interpreting charts, not just creating them

Why This Topic Matters

Raw tables are important, but they can be difficult to interpret quickly. Visualisations help you:

• compare categories more easily
• spot trends and patterns
• understand distributions
• identify unusual values or outliers
• communicate findings to other people

A good chart can make an important pattern visible in seconds. A strong analyst knows how to create that chart and explain what it means.

What Is Data Visualisation?

Data visualisation is the process of representing data using visual forms such as charts, graphs, and plots. Instead of reading hundreds of rows one by one, you can use a chart to quickly understand:

• which category is highest or lowest
• whether values are increasing or decreasing
• whether most values are clustered together
• whether two variables seem related

Visualisation turns data into something easier to explore and easier to explain.

What Questions Can Charts Answer?

Most beginner visualisations are built to answer one of four common analytical questions:

• Comparison — How do categories differ?
• Trend — How do values change over time or sequence?
• Distribution — How are values spread out?
• Relationship — How do two variables move together?

These four question types will guide much of this module. As you learn each chart type, you should always ask: What question is this chart helping me answer?

Comparison

A comparison chart helps you compare values across categories. Examples:

• Which gender has the higher average mathematics score?
• Which lunch type has the higher average reading score?
• Which race/ethnicity group appears most often in the dataset?

Bar charts are commonly used for comparison because they make it easy to compare category heights. When reading a comparison chart, pay attention to:

• which category is highest
• which category is lowest
• whether the differences are large or small

Trend

A trend chart helps you see how data changes over time or across an ordered sequence. Examples:

• How do monthly sales change over a year?
• Does attendance increase or decrease over time?
• Are scores improving across multiple test dates?

Line charts are commonly used for trends because they show movement across an ordered axis. When reading a trend chart, ask:

• Is the trend increasing, decreasing, or fluctuating?
• Are there sharp changes?
• Are there periods of stability?

Distribution

A distribution chart helps you understand how values are spread. Examples:

• Are most mathematics scores clustered in the middle?
• Are scores spread widely or tightly?
• Are there unusually low or high values?

Histograms and boxplots are useful for distribution. When reading a distribution chart, ask:

• Where are most values concentrated?
• Is the spread narrow or wide?
• Are there outliers?

Relationship

A relationship chart helps you explore how two numerical variables behave together. Examples:

• Do reading and writing scores increase together?
• Is there a connection between hours studied and exam score?
• Do two measurements seem positively related, negatively related, or unrelated?

Scatter plots are especially useful here. When reading a relationship chart, ask:

• Do the points move upward, downward, or randomly?
• Is the relationship strong or weak?
• Are there any unusual points far from the rest?

Reading Charts as an Analyst

Creating a plot is only part of the job. You also need to read and interpret what the chart shows. A simple way to do that is to follow this process:

1. Read the title. What is the chart about?
2. Look at the axes. What do they represent?
3. Identify the chart type. Is it showing comparison, trend, distribution, or relationship?
4. Look for the main pattern. What stands out first?
5. Look for unusual values. Are there outliers, gaps, or surprising differences?
6. State the message in words. What does the chart suggest?

This habit will help you move from making plots to doing real analysis.

Example: Interpreting a Simple Comparison

Imagine a bar chart showing average mathematics score by gender. You might describe it like this:

• the male group has a higher average mathematics score than the female group
• the difference is visible, but it should still be checked numerically
• this chart supports a comparison question, not a relationship or trend question

Notice that the goal is not just to say “this is a bar chart.” The goal is to explain what the chart communicates.

Example: Interpreting a Distribution

Imagine a histogram of reading scores. A useful interpretation might be:

• most scores fall in the middle range
• very low and very high scores are less common
• the histogram helps show how the reading scores are distributed across students

Again, the value of the chart comes from interpretation, not just from generating it.

Tools you’ll use

You will create charts in Google Colab using:

• Matplotlib for basic plots and detailed control.
• Seaborn (later) for statistical visualisations built on top of Matplotlib.
• Plotly Express (optional) for interactive charts.

All examples will start from Pandas DataFrames, often using the StudentsPerformance dataset you worked with in the Working with Data module.

Choosing the Right Kind of Chart

One of the most common mistakes in data visualisation is choosing a chart that does not match the question. For example:

• use a bar chart to compare categories
• use a line chart to show change over time
• use a histogram to study the distribution of one numerical variable
• use a scatter plot to study the relationship between two numerical variables

Good analysts do not start by asking, “Which chart looks nice?” They start by asking, “What am I trying to show?”

Google Colab Activity

Open your Students Performance dataset in Google Colab. Then answer these questions in writing before making any charts:

1. Which question would require a comparison chart?
2. Which question would require a distribution chart?
3. Which question would require a relationship chart?
4. Which question would require a trend chart if your dataset included dates?

This activity will help you think about chart selection before learning plotting syntax.

Practice Exercise

For each question below, decide which type of chart would be most appropriate:

1. Which parental education group has the highest average writing score?
2. How are mathematics scores distributed across students?
3. Is there a relationship between reading score and writing score?
4. How do average scores change across semesters?
5. Which lunch category appears most often?

Write the chart type you would choose and explain why.

Self Evaluation

Check Your Understanding

1. What is the main purpose of data visualisation?

To replace datasets completelyTo make data easier to explore, understand, and communicateTo avoid using Python codeTo remove the need for summary statistics

2. Which analytical question is most closely related to a bar chart?

How are values distributed?How do two numerical variables relate?How do categories compare?How does a value change continuously over time?

3. Which chart type is most suitable for studying the distribution of one numerical variable?

HistogramLine chartScatter plotPie chart

4. When interpreting a chart, what should you check first?

The background colorThe title and axesThe notebook themeThe Python version

5. A scatter plot is most useful for answering which type of question?

How categories compareHow a single variable is distributedHow two numerical variables are relatedHow file formats differ

Challenge Exercise

Imagine a classmate creates four charts from the same dataset:

• a bar chart
• a line chart
• a histogram
• a scatter plot

For each chart, write:

• what question it is best suited to answer
• what kind of pattern you would look for
• one mistake someone might make when interpreting it

This exercise is designed to strengthen both chart selection and chart reading.

Common Mistakes

Treating Charts as Decoration

Charts are not just for making notebooks look better. They are analytical tools that help you understand and communicate data.

Choosing a Chart Before Defining the Question

If you do not know what you are trying to show, it is easy to choose the wrong visual. Always start with the analytical question.

Ignoring Axes and Labels

A chart cannot be interpreted correctly if you do not read its axes, title, and labels carefully.

Describing the Chart Without Interpreting It

Saying “this is a histogram” is not enough. A stronger answer explains what the chart suggests about the data.

Key Takeaways

In this lesson, you learned that:

• data visualisation helps turn raw data into understandable patterns
• charts often answer one of four question types: comparison, trend, distribution, or relationship
• reading a chart requires attention to titles, axes, patterns, and unusual values
• good analysts know how to interpret a chart, not just create one

Continue Your Journey

Next, you will learn Matplotlib Basics, where you will create your first charts in Python and start turning these ideas into actual visualisations in Google Colab.