Scatter graphs plot two variables to see if there is a relationship (correlation) between them. You need to describe the type and strength of correlation and draw lines of best fit.
Put your revision into practice
Choose a topic and try a short quiz with instant explanations. Free, with no signup.
Revising scatter graphs & correlation the night before an exam? These are condensed revision notes — the facts, formulas and mistakes examiners actually mark, with no long explanations. Prefer a full walkthrough with step-by-step reasoning instead? Search “Scatter graphs & correlation GCSE” on the site for the complete guide.
🔑
Key facts to remember
1Positive correlation: as one variable increases, the other increases.
2Negative correlation: as one variable increases, the other decreases.
3No correlation: no clear pattern between the variables.
4Correlation can be strong (points close to a line) or weak (points scattered widely).
5Correlation does not imply causation — two things may be related without one causing the other.
6An outlier is a point that does not fit the general pattern.
✍️
Worked examples
Example 1
A scatter graph of temperature vs ice cream sales shows points rising from bottom-left to top-right. Describe the correlation.
Working
As temperature increases, ice cream sales increase
Points rise from bottom-left to top-right → positive correlation
Points are close together → strong positive correlation
AnswerStrong positive correlation
Example 2
A scatter graph shows hours of revision vs exam errors. Describe the expected correlation.
Working
More revision → fewer errors
One increases while the other decreases → negative correlation
AnswerNegative correlation
⚠️
Common mistakes
✗Confusing positive and negative correlation — positive means both go up together.
✗Stating that a correlation proves one variable causes the other.
✗Joining plotted points with a line instead of drawing a line of best fit.
🎯
Exam tips
✓Use precise language: "strong positive correlation" not just "positive".
✓Correlation describes the relationship, not causation — avoid saying one variable "causes" the other.
Ready to test yourself on Scatter graphs & correlation?
Get AI-marked practice questions on exactly this subtopic.