Scatter Graphs & Correlation
Studying this course? Keep your progress in one place.
Key facts
- A scatter graph plots pairs of values as single points, drawn as small crosses.
- Positive correlation: both increase. Negative: one rises as the other falls. Zero: no clear pattern.
- A line of best fit is one ruled straight line across the data with points roughly balanced either side.
- Interpolation (inside the data range) is reliable; extrapolation (outside it) is not.
- Correlation does not prove that one variable causes the other.
Scatter graphs
A scatter graph shows whether two variables are related, with one point for each pair.
A scatter graph plots pairs of values, one variable on each axis, to see whether a relationship exists. Each pair of data is a single point. In exams, plot points as small, neat crosses (×) so the exact position is clear.
Each pair of values on a scatter graph is shown as what?
Correlation
Correlation describes how the points trend, and how closely they follow a straight line.
Correlation describes the relationship shown by the pattern of points. The closer the points are to a straight line, the stronger the correlation; the more scattered, the weaker. Correlation shows a link exists, but does not prove that one variable causes the other.
Positive
- Points trend upward
- Both variables increase
Negative
- Points trend downward
- One rises as the other falls
Zero
- No clear pattern
- No linear relationship
As the temperature rises, ice cream sales rise. What type of correlation is this?
Line of best fit
Draw one ruled straight line along the trend, with points balanced on both sides.
A line of best fit is drawn by eye to represent the trend. Use a ruler and draw a single straight line (not a curve or joined segments). It should run across the full range of the data, with points roughly evenly spread above and below. It does not have to pass through the origin or any particular point. There is no single correct line, but it must follow the trend.
Which description of a line of best fit is correct?
Making predictions
Read from the line of best fit, and say whether the prediction is interpolation or extrapolation.
Find the known value on one axis, go to the line of best fit, then read across or down to the other axis for an estimate. Interpolation is estimating inside the range of the data and is generally reliable. Extrapolation is estimating outside it, which is less reliable because the trend may not continue.
Worked example
Use the line to estimate the score after 5.5 hours and after 10 hours, and comment on reliability.
- 1
At 5.5 hours: . This is inside the data range (1 to 8 hours): interpolation, reliable.
- 2
At 10 hours: . This is outside the range: extrapolation, unreliable. A score above 100 may not even be possible.
The data covers heights from 150 cm to 180 cm. You predict a value at 195 cm. What is this called?
Try an exam question
A scatter graph shows hours studied (x) and test score (y). The line of best fit is and the data covers 1 to 8 hours. (a) What type of correlation is shown? (b) Use the line to estimate the score for 6 hours. (c) Explain why an estimate for 12 hours may be unreliable.
[4 marks]
- [1]Positive correlation
- [1]
- [1]72
- [1]12 hours is outside the data range (extrapolation), so the trend may not continue
That's the notes covered.
Carry on to the next subtopic.
Try a question
What type of correlation does this data show?
1 mark, marked against the mark scheme
What type of correlation does this data show?
1 mark, marked against the mark scheme