English:Data Visualisation
Data Visualisation
Introduction
Data visualisation means representing data with visual forms such as bars, lines, points, areas, symbols, colours, and maps. A good visualisation helps you notice patterns, compare values, ask better questions, and communicate evidence clearly. A poor visualisation can confuse you or even make small differences look much larger than they really are.
In this aiMOOC for Grades 7–8, you will learn to read, question, choose, and create common charts. You will also explore how visual design affects meaning and how to spot graphs that may mislead you.
Why Visualise Data?
A table is useful when you need an exact value. A chart is often better when you want to see a pattern quickly. For example, a line chart can make a rise or fall over time easier to notice, while a bar chart makes category comparisons easier.
Data visualisation is part of statistics, mathematics, information technology, science, geography, journalism, and many other fields. The same basic questions apply everywhere: What data are shown? How are they encoded? What pattern can you see? What evidence supports your conclusion?
The Data Behind the Picture
Before choosing a chart, identify what kind of data you have. Categorical data place observations into groups, such as favourite school subject or transport type. Numerical data describe quantities, such as height, temperature, distance, or time.
Also ask whether your values are counts, percentages, measurements, or rates. Units matter. A graph of distance without kilometres or miles is incomplete, and a percentage should usually make clear what whole or group it refers to.
A visualisation is not separate from its data. If the data are inaccurate, incomplete, biased, or badly collected, a beautiful chart can still give a false impression.
Choosing a Chart
Bar charts compare amounts across categories. Bars should normally have equal widths and a common baseline so that lengths can be compared fairly.
Line charts are useful for showing change across ordered values, especially time. Connecting points suggests continuity, so a line chart is usually not the best choice for unrelated categories.
Pie charts show parts of a whole. They work best when there are only a few categories and when the parts add to the complete whole. A bar chart is often easier when you need precise comparisons.
Histograms show the distribution of numerical data by grouping values into intervals. Unlike a bar chart for categories, neighbouring bars in a histogram represent connected numerical intervals.
Scatter plots show pairs of numerical values. Each point represents one observation, and the overall cloud of points can suggest a positive relationship, a negative relationship, or little visible relationship.
Reading a Visualisation Like a Detective
Do not begin by staring only at the tallest bar or most dramatic colour. Read the structure first. Check the title, axes, labels, units, legend, time period, data source, and scale. Then identify the largest and smallest values, patterns, clusters, gaps, changes, and unusual points.
An outlier is a value that lies far from most of the other values. An outlier can be important, but it is not automatically an error. You should investigate possible reasons before removing or ignoring it.
A trend is an overall direction in the data. A trend does not mean that every single value follows the direction perfectly.
Correlation Is Not Causation
A scatter plot may show that two variables change together. This is called correlation. However, correlation alone does not prove that one variable causes the other. A third factor may influence both, or the pattern may have another explanation.
For example, if two quantities both increase during summer, that does not automatically mean one caused the other. Good data reasoning separates what the graph shows from what you think might explain it.
How Graphs Can Mislead
A chart can be technically based on real numbers but still create a distorted impression. Warning signs include a cut-off axis, unequal intervals, missing units, selective time ranges, unnecessary 3D effects, confusing colours, or categories chosen to hide an important comparison.
Compare these two bar-chart examples. The first uses a non-truncated scale. The second begins its vertical scale above zero, which makes the difference in bar lengths look much more dramatic.
For bar charts, bar length is the main visual signal. Because viewers compare those lengths, a truncated vertical axis can strongly exaggerate differences. Other chart types may sometimes use non-zero baselines for valid reasons, but the scale must still be clear and appropriate.
Design for Clarity and Accessibility
A good chart makes the important comparison easy to see. Use a clear title, readable labels, suitable units, sensible scales, and only as much decoration as the message needs. Avoid turning a chart into a puzzle.
Colour can help group information, highlight a key value, or show ordered intensity. However, do not use colour as the only way to communicate meaning. Labels, shapes, patterns, or direct annotations can make a chart easier to understand for people with colour-vision differences and for readers using different screens or printed copies.
When possible, write a short text description of the main message. Accessibility is not an extra feature; it is part of clear communication.
Data Stories and Historical Examples
Visualisation can help people notice patterns that are difficult to see in a table. In 1854, physician John Snow mapped cholera deaths in London. The clusters on the map helped him investigate the relationship between cases and local water sources. The map is an early example of using spatial data to study a public-health problem.
In the 1850s, Florence Nightingale used statistical diagrams to communicate causes of mortality among British soldiers. Her polar-area diagram combined data and visual design to support an argument about preventable deaths and sanitation.
These historical examples show that data visualisation is not only about making information attractive. It can support investigation, explanation, and decisions.
Data, Infographics, and Digital Media
Online charts and infographics can combine data with text, icons, maps, and images. This can make a story memorable, but it also increases the number of design choices that can influence your interpretation.
When you meet a chart online, ask who made it, where the data came from, when the data were collected, whether important context is missing, and whether the visual design matches the numbers. A trustworthy-looking graphic is not automatically trustworthy evidence.
A Simple Workflow for Creating Your Own Visualisation
Start with a question you can answer with data. Collect or find suitable data and check what each value means. Clean obvious mistakes carefully without changing results simply because they look inconvenient. Choose a chart type that matches the question. Build the chart with clear labels, units, and scales. Then test it: ask another person what message they see before you explain it.
Finally, revise. Good visualisation is an editing process. Remove clutter, correct confusing labels, add needed context, and make sure the chart supports a truthful interpretation of the data.
Interactive Tasks
Quiz: Test Your Knowledge
Which chart is usually best for comparing amounts across separate categories? (Bar chart) (!Line chart) (!Scatter plot) (!Histogram)
Which chart is usually best for showing change over time? (Line chart) (!Pie chart) (!Pictogram) (!Bar chart)
What does a histogram usually show? (Numerical values grouped into intervals) (!Separate categories arranged alphabetically) (!Only percentages of one whole) (!Locations on a map)
What is the main purpose of a scatter plot? (To show the relationship between two numerical variables) (!To show parts of a single whole) (!To list exact values in rows) (!To arrange events into categories)
When is a pie chart most suitable? (When a few categories form parts of one whole) (!When hundreds of values must be compared precisely) (!When changes over many years are the main focus) (!When two numerical variables are being related)
Why can a truncated vertical axis make a bar chart misleading? (It can exaggerate differences in bar lengths) (!It always changes the original data values) (!It turns numerical data into categories) (!It prevents a chart from having a title)
Which feature helps a reader understand what numbers on an axis mean? (A clear scale with labels and units) (!Decorative background images) (!A larger company logo) (!Three dimensional effects)
What does correlation by itself prove? (That two variables show a relationship in the observed data) (!That one variable definitely causes the other) (!That every point follows the same pattern) (!That the data contain no outliers)
Which design choice improves accessibility? (Use labels or patterns as well as colour) (!Use colour as the only signal) (!Remove all text from the chart) (!Make every category the same symbol without labels)
What did John Snow use a map to investigate? (The spatial pattern of cholera cases) (!The brightness of stars) (!The speed of trains) (!The height of mountains)
Memory Game
| Bar chart | Compares amounts across categories |
| Line chart | Shows change across ordered values such as time |
| Histogram | Groups numerical data into intervals |
| Scatter plot | Shows a relationship between two numerical variables |
| Legend | Explains visual symbols or colours |
| Outlier | A value far from most other values |
Drag and Drop
| Match the correct terms. | Topic |
|---|---|
| Bar chart | Compare separate categories |
| Line chart | Show change over time |
| Histogram | Show a numerical distribution |
| Scatter plot | Explore a relationship between two numerical variables |
| Pie chart | Show a few parts of one whole |
...
Crossword Puzzle
| Histogram | Which chart groups numerical values into intervals? |
| Scatterplot | Which chart uses points to show pairs of numerical values? |
| Legend | What explains colours or symbols in a chart? |
| Outlier | What is a value far from most of the other values called? |
| Scale | What shows how positions on an axis correspond to values? |
| Correlation | What word describes a relationship between two variables? |
LearningApps
Cloze Text
Open-Ended Tasks
Easy
- Chart Hunt: Find three charts in books, websites, or newspapers. Identify each chart type and write one sentence about what it communicates.
- Classroom Survey: Ask classmates one simple categorical question, record the results, and create a labelled bar chart.
- Chart Makeover: Redraw a cluttered or confusing chart so that the title, labels, scale, and main comparison are clearer.
- Visual Diary: Keep a one-week record of one numerical quantity, such as daily reading minutes, and display it in a suitable chart.
Standard
- School Travel Survey: Collect anonymous data about how students travel to school, choose a suitable visualisation, and explain why your chart type fits the data.
- Misleading Graph Detective: Find a real graph that could mislead a reader, identify the design choice causing the problem, and create a fairer version.
- Infographic Story: Create a one-page infographic from a small dataset, combining a chart with concise explanatory text and a clear source note.
- Chart Comparison: Show the same dataset in two different chart types and write a comparison explaining which version communicates the main message more effectively.
Advanced
- Data Journalism: Investigate a question about your school or community using reliable data, create two visualisations, and write a short evidence-based news report.
- Accessibility Audit: Review three charts for readability and accessibility, then redesign one so that meaning does not depend on colour alone.
- Local Data Map: Collect or find location-based data for your community, create a simple map-based visualisation, and explain one pattern and one limitation.
- Visualisation Debate: Prepare a short presentation arguing whether a chosen chart is fair and effective, using evidence about scales, labels, context, and chart type.
Learning Assessment
- Chart Selection: Given three different datasets, choose an appropriate visualisation for each and justify your choices using the kind of data and the question being asked.
- Critical Reading: Analyse a graph with a missing label, unusual scale, or selective time range and explain how the design affects interpretation.
- Evidence and Claims: Write two claims supported by a supplied chart and one claim that the chart does not support, explaining the difference.
- Correlation Reasoning: Interpret a scatter plot, describe its visible relationship, and propose two possible explanations without claiming causation from correlation alone.
- Design Revision: Improve a weak visualisation and annotate at least four changes that make it clearer, fairer, or more accessible.
- Transfer Task: Use data from science, geography, sport, or another subject to create a visualisation and explain how it helps answer a subject-specific question.
Evidence of Learning
Knowledge: You can explain the purposes of bar charts, line charts, pie charts, histograms, scatter plots, legends, scales, and labels, and you can distinguish categorical from numerical data.
Skills: You can read values accurately, identify patterns and outliers, choose suitable chart types, question scales and sources, and separate correlation from causation.
Products: Your work may include labelled charts, a revised misleading graph, a survey visualisation, an infographic, a map, or a short data story.
Transfer: You can apply visualisation skills to mathematics, science, geography, information technology, media literacy, and everyday claims involving data.
Communication: You can explain what a chart shows, what it does not show, and which design decisions help a reader interpret the evidence fairly.
OERs on the Topic
Linked Learning Areas