Data & Statistics · Lesson 7 of 14
Charts and Graphs
A good chart answers a question at a glance. The skill is matching the chart type to the kind of data and the question asked.
What you will learn
- Choose the right chart for the question.
- Read pie charts, line graphs and scatter graphs.
- Spot charts that mislead.
Before you start: you should know how to read bar charts, line graphs and pie charts.
The idea
Raw data is hard to take in. A chart turns a table of numbers into a picture that shows patterns immediately: which category is biggest, how something changed over time, or whether two things are related.
Which chart for which job?
| Question | Chart | Data type |
|---|---|---|
| How do categories compare? | Bar chart | Categories (e.g. favourite fruit) |
| What share of the whole is each part? | Pie chart | Parts of a single total |
| How did it change over time? | Line graph | Continuous time series |
| How is a measurement distributed? | Histogram | Grouped continuous data (e.g. heights) |
| Are two variables related? | Scatter graph | Pairs of measurements |
Bar charts vs histograms
They look similar but answer different questions. Bar charts show separate categories, so the bars have gaps and the order can change. Histograms show continuous ranges, so the bars touch, and it is the area of each bar that represents frequency.
Scatter graphs and correlation
On a scatter graph each point is one individual measured on two variables. If points drift upward from left to right, there is positive correlation; downward means negative correlation; a shapeless cloud means no correlation. A line of best fit summarises the trend and can be used to estimate values.
Worked examples
Example 1: Drawing a pie chart
A survey of students’ travel to school: walk , bus , car , cycle . Find the pie-chart angles.
- Each student gets .
- Walk: . Bus: .
- Car: . Cycle: .
- Check: . ✓
Answer: , , ,
Example 2: Reading a line of best fit
A line of best fit for hours revised () against test score () is . Estimate the score for hours of revision, and comment on using it for hours.
- Substitute : .
- For hours the line predicts , which is impossible if the test is out of .
Answer: about . Predicting far outside the data (extrapolation) is unreliable.
Example 3: Spotting a misleading graph
A bar chart shows sales of and units, but the vertical axis starts at . The second bar looks four times taller than the first.
- Actual increase: units.
- As a percentage: .
Answer: a modest rise made to look dramatic by a truncated axis.
Tip: correlation is not causation. Ice-cream sales and sunburn cases rise together, but ice cream does not cause sunburn. Hot weather drives both.
Why it matters
Charts are how most people meet data: in news reports, company presentations, fitness apps, energy bills and public health briefings. Being able to read them accurately and spot tricks like truncated axes, 3D distortion or cherry-picked time ranges is an essential part of being an informed citizen.
Practice
Your turn. Type or choose an answer, then press Check. Getting one wrong is fine, that is what practice is for. You can try as many times as you like.