Learning outcomes
- detect misleading categorical graphs
- explain why scale distortion is dangerous
- check whether percentages and counts are presented honestly
- present categorical data responsibly
Misleading graphs
A graph is misleading when it creates an incorrect visual impression. Common causes:- truncated axis without explanation
- non-uniform scaling
- 3D effects that distort area
- unequal bar widths
- percentages that do not sum correctly
Truncated axes
- Sometimes the vertical axis starts above zero.
- This may exaggerate small differences.
- starting at zero is often the safest and clearest choice
Counts vs percentages
- Counts show actual numbers.
- Percentages show share of the total.
- comparing raw counts from one dataset with percentages from another without context
Too much decoration
- 3D bars
- shadows
- unnecessary colours
- pictographs that scale icons by area
Responsible presentation checklist
- Is the scale clear?
- Do the bars match the values?
- Are percentages correct?
- Is the total context clear?
- Is the design simple enough to read?
Exam hints and traps
- If a bar chart exaggerates a tiny difference visually, suspect scale distortion.
- Bigger-looking bars do not matter unless scale supports them.
- Pie charts with too many slices are hard to read.
- Counts and percentages answer different questions.
Quick practice
- Why can a truncated axis mislead?
- Why are 3D bar charts risky in statistics?
- When should percentages be preferred over counts?
Answer key
- It can make small differences appear much larger than they are.
- Visual distortion can make comparison less accurate.
- When comparing category share within a total, especially across groups of different sizes.