Scatter Graphs and Line of Best Fit
A scatter graph plots pairs of values for two related variables as points on a coordinate grid, so that any pattern or relationship (correlation) between them can be seen. In GCSE Statistics, a line of best fit is a straight line drawn through the middle of the points on a scatter graph, used to describe and estimate values for the relationship between the two variables.
Method
- Plot each pair of values as a single point on a coordinate grid, with one variable on the horizontal axis and the other on the vertical axis.
- Look at the overall pattern of the points to decide whether there is positive, negative or no correlation.
- If the points show a roughly linear pattern, draw a single straight line of best fit that passes as close as possible to all the points, with a similar number of points on each side.
- The line of best fit does not have to pass through the origin or through any particular point, unless the data specifically supports this.
- Use the line of best fit to estimate an unknown value by reading up or across from a known value to the line, then across or down to the other axis.
- Be cautious about extrapolating, estimating far beyond the range of the plotted data, since the relationship may not continue in the same way.
Common mistakes
MistakeCalling a relationship strong merely because it is positive, without considering how closely the points follow a pattern.
InsteadPositive or negative describes direction. Judge strength from how tightly the points cluster around a clear pattern, not from the direction alone.
MistakeForcing a line of best fit through the origin or through every plotted point, even when the pattern does not support that choice.
InsteadDraw one straight line through the middle of the points, with a broadly balanced spread on each side, unless the data clearly requires a particular intercept.
MistakeTreating an estimate far outside the plotted range as reliable because the line gives a numerical answer.
InsteadIdentify extrapolation and qualify it: the relationship may change beyond the observed range, so an estimate inside the data range is usually more dependable.
Worked examples
Example 1
A scatter graph plots the number of hours revised against the test score for 10 students. A line of best fit is drawn. A student who revised for 5 hours reads across to the line of best fit at a test score of 62, and a student who revised for 8 hours reads across to the line of best fit at a test score of 80. Use these two points on the line to estimate the test score for a student who revises for 6.5 hours.
- Find how much the score increases for each extra hour of revision along the line: (80 - 62)/(8 - 5) = 18/3 = 6 marks per hour.
- Find the difference in hours between the student we want and the known point at 5 hours: 6.5 - 5 = 1.5 hours.
- Multiply the increase per hour by this difference: 1.5 x 6 = 9 marks.
- Add this to the score at 5 hours: 62 + 9 = 71.
- Final answer: the estimated test score for 6.5 hours of revision is 71.
Example 2
A line of best fit for two related variables passes through the points (4, 18) and (10, 42). Use the line to estimate the second variable when the first is 7.5, and estimate the first variable when the second is 30. State whether each estimate is interpolation or extrapolation.
- Gradient = (42 - 18) / (10 - 4) = 24 / 6 = 4 units of the second variable per unit of the first.
- For a first-variable value of 7.5, the change from 4 is 3.5, so the estimated second variable is 18 + 3.5 x 4 = 32.
- For a second-variable value of 30, the change from 18 is 12, so the estimated first variable is 4 + 12 / 4 = 7.
- Both requested values lie between the supplied coordinates, so both estimates are interpolation rather than extrapolation.
- Final answer: the estimates are 32 when the first variable is 7.5 and 7 when the second variable is 30; both are interpolation estimates.
Practice questions
Q1[1 mark]A scatter graph shows points that generally rise from bottom left to top right. What type of correlation does this show?Show answer
Answer: Positive correlation.
Q2[1 mark]A scatter graph shows points that generally fall from top left to bottom right. What type of correlation does this show?Show answer
Answer: Negative correlation.
Q3[1 mark]A scatter graph shows points scattered with no clear pattern. What type of correlation does this show?Show answer
Answer: No correlation (zero correlation).
Q4[1 mark]A line of best fit passes through the points (0, 4) and (10, 24). Find its gradient.Show answer
Answer: 2 (rise/run = (24-4)/(10-0) = 20/10 = 2).
Q5[1 mark]Using the line of best fit from the previous question, estimate the y-value when x = 6.Show answer
Answer: 16 (y = 4 + 2 x 6 = 16, using the gradient of 2 and starting value of 4 at x = 0).
Q6[2 marks]Explain why estimating a y-value for x = 50 would be unreliable if all the plotted data only covers x-values from 0 to 20.Show answer
Answer: This would be extrapolation, estimating far outside the range of the collected data, where the relationship might not continue in the same pattern.
Exam-style questions
Written in the style of a GCSE Statistics exam paper, with a full mark scheme.
Exam tips
- Name the variables on the axes before reading a point from the line, and include units in the estimate when the question gives them.
- Use the line of best fit rather than a nearby raw point, then read across or down carefully and state sensible precision.
- Say when an estimate is extrapolation because it lies outside the observed range, and explain why that reduces confidence.
Q1[2 marks]A scatter graph plots the age of a car, in years, against its value, in pounds. Describe the type of correlation you would expect to see, and explain your reasoning.Show mark scheme
Tick each line you got.
Nothing ticked yet - 2 available
Q2[3 marks]A scatter graph shows the number of ice creams sold plotted against the daily temperature, in degrees C, for 12 days. A line of best fit is drawn. On the line, a temperature of 20 degrees C corresponds to 45 ice creams sold, and a temperature of 30 degrees C corresponds to 85 ice creams sold. Use the line of best fit to estimate the number of ice creams sold when the temperature is 25 degrees C.Show mark scheme
Tick each line you got.
Nothing ticked yet - 3 available
Q3[3 marks]A student says: 'My scatter graph shows a strong positive correlation between the number of pet cats owned in a town and the number of trees in that town, so owning more cats must cause more trees to grow.' Explain why the student's conclusion is not justified.Show mark scheme
Tick each line you got.
Nothing ticked yet - 3 available
See real GCSE Statistics past-paper questions, with official mark schemes
This topic in real past papers
We have tagged some boards' past papers question by question (not every board or year yet). This topic comes up in:
- Edexcel GCSE Statistics past papers: 12 questions across 10 papers, latest tagged June 2023
- AQA GCSE Statistics past papers: 11 questions across 9 papers, latest tagged June 2023
Free printable worksheet
Want more practice on paper? Download the Scatter Graphs and Line of Best Fit worksheet pack - 32 pages of exam-style questions with a full mark scheme. Free to download, no card or account. Print it for personal and classroom use.
This topic is chapter 12 of GCSE Statistics Foundation Essentials 2 and chapter 15 of GCSE Statistics Higher Essentials 2, the whole course as one free printable PDF. Its worksheet is also chapter 4 of GCSE Statistics Foundation Topic Workbook 5 and chapter 7 of GCSE Statistics Higher Topic Workbook 6, free printable practice books.
Other versions of this worksheet: