Scatter Graphs GCSE Maths Revision Guide
Revise scatter graphs GCSE Maths with clear methods, worked examples, correlation, lines of best fit, common mistakes and exam questions.

Scatter Graphs GCSE Maths Revision Guide
Scatter graphs compare two numerical variables to show whether a relationship exists between them. For scatter graphs GCSE Maths questions, you need to plot paired data, describe correlation, identify outliers, draw a line of best fit and use it to estimate values.
This guide explains each skill step by step. It also includes scatter graphs worked examples and exam-style questions with answers suitable for GCSE Maths students.
What is a scatter graph?
A scatter graph, also called a scatter diagram, displays pairs of numerical data. Each pair is plotted as one point.
For example, a teacher could record each student's revision time and test mark. A student who revised for 444 hours and scored 626262 marks would be represented by the point (4,62)(4,62)(4,62).
Scatter graphs are appropriate when both variables are numerical, such as:
- height and mass
- temperature and ice cream sales
- revision time and test score
- age of a car and its value
They are not normally suitable for comparing categories such as favourite colours or methods of transport.
The independent or explanatory variable is usually placed on the horizontal axis. The dependent or response variable is usually placed on the vertical axis because its value may be related to the horizontal variable. In some GCSE questions, you will simply be given the axis labels and do not need to classify the variables.
How to plot a scatter graph
Follow this method when constructing a scatter graph from a table of values.
- Read both axis labels and units carefully.
- Work out the value represented by each small grid interval.
- Take one pair of values from the table.
- Move across to the first value and then up to the second value.
- Mark the point accurately with a small cross.
- Repeat for every pair and check that no rows have been missed.
Do not join the points. A scatter graph shows separate observations rather than a continuous sequence.
Worked example 1: plotting paired data
A student's revision time is 555 hours and their test score is 686868 marks. Revision time is shown on the horizontal axis and test score on the vertical axis.
The coordinate is:
(5,68)(5,68)(5,68)Start at 555 on the horizontal axis, move vertically to 686868 on the test-score axis and plot a small cross. Reversing the coordinate to (68,5)(68,5)(68,5) would plot the variables incorrectly.
Correlation on scatter graphs
Correlation describes the direction and strength of the relationship between two variables. At GCSE, you should recognise positive correlation, negative correlation and no correlation.
| Type of correlation | Pattern of points | Interpretation |
|---|---|---|
| Positive correlation | Points generally rise from left to right | As one variable increases, the other tends to increase |
| Negative correlation | Points generally fall from left to right | As one variable increases, the other tends to decrease |
| No correlation | No clear upward or downward pattern | There is no evident relationship between the variables |
The closer the points lie to a clear trend, the stronger the correlation. Points that are spread widely around the trend indicate weaker correlation.
You may therefore describe a relationship as:
- strong positive correlation
- weak positive correlation
- strong negative correlation
- weak negative correlation
- no correlation
Unless the question asks about strength, stating the correct direction may be sufficient. If the question asks you to describe the relationship, give both its direction and its meaning in context.
Worked example 2: interpreting correlation
A scatter graph compares the age of several cars with their selling prices. The points generally slope downwards from left to right.
This shows negative correlation. In context, older cars tend to have lower selling prices.
Use cautious language such as tends to. Correlation describes a general pattern, so it does not mean every older car must be cheaper than every newer car.
Correlation does not prove causation
A correlation shows that two variables are associated. It does not, by itself, prove that changing one variable causes the other to change.
For example, sunglasses sales and ice cream sales may have positive correlation. Buying sunglasses does not cause people to buy ice cream. Hot, sunny weather can affect both variables.
A third variable that helps explain an apparent relationship is sometimes called a confounding variable. In GCSE answers, the key point is usually that correlation does not necessarily mean causation.
Lines of best fit
A line of best fit is a straight line that represents the overall trend of the points. It can be used to make estimates.
When drawing one by eye:
- use a ruler
- follow the overall direction of the data
- place the line through the middle of the cluster
- aim for a reasonable balance of points above and below it
- do not force the line through the origin
- do not join individual points
- ignore an outlier when it clearly does not follow the general pattern
There is not always one unique line that must be drawn. Slightly different sensible lines can be accepted, and estimates may therefore fall within an acceptable range.
A line of best fit does not usually need to pass through any particular data point. It represents the whole trend, not one observation.
Worked example 3: using a line of best fit
A scatter graph compares hours of revision with test marks. To estimate the mark for a student who revises for 666 hours:
- Locate 666 on the revision-time axis.
- Draw or trace a vertical line to the line of best fit.
- From that intersection, move horizontally to the test-mark axis.
- Read the estimated mark from the scale.
Suppose the line gives approximately 747474 marks. The answer should be presented as an estimate:
Estimated mark≈74\text{Estimated mark} \approx 74Estimated mark≈74To estimate revision time from a given mark, reverse the process. Begin at the vertical axis, move horizontally to the line of best fit, and then move down to the horizontal axis.
Interpolation and extrapolation
The position of an estimate matters.
Interpolation means estimating within the range of the observed data. Extrapolation means estimating outside that range.
Suppose the recorded revision times range from 111 to 888 hours:
- estimating a score for 555 hours is interpolation
- estimating a score for 121212 hours is extrapolation
Interpolation is generally more reliable because the estimate lies among the observed values. Extrapolation is less reliable because the existing pattern may not continue beyond the data range.
If an exam asks why a prediction is unreliable, check whether it involves extrapolation. A suitable answer is that the prediction is outside the range of the collected data, so there is no evidence that the trend continues that far.
Outliers and unusual points
An outlier is a point that lies noticeably away from the general pattern of the other points. At GCSE, you may be asked to identify an outlier or suggest a reason for it.
Possible explanations depend on the context. An outlier could result from:
- a measurement or recording error
- unusual circumstances
- a genuine observation that differs from the general trend
Do not automatically state that an outlier is an error. It may be valid data. If asked to identify one, give its coordinate or describe the relevant observation precisely.
An outlier can affect the apparent correlation. It can also distort a line of best fit if the line is drawn towards that single point rather than through the main cluster.
Scatter graphs worked example: a complete interpretation
A company records the outside temperature and the number of hot drinks sold on eight days. The scatter graph has a strong downward trend, with one point far from the main cluster.
A complete interpretation could include:
- There is strong negative correlation.
- As temperature increases, the number of hot drinks sold tends to decrease.
- The isolated point is an outlier.
- A line of best fit should follow the main cluster rather than being pulled strongly towards the outlier.
- An estimate inside the recorded temperature range is interpolation and is more reliable than an estimate outside it.
Notice that the contextual statement names both variables. Simply writing “one goes down as the other goes up” is less precise.
Common mistakes in scatter graphs exam questions
Reversing coordinates
Always use the horizontal value first. A pair written as (x,y)(x,y)(x,y) means move along the horizontal axis to xxx, then up or down to yyy.
Misreading the scale
Do not assume each small square represents 111. Use two labelled values to determine the size of each interval before plotting or estimating.
Joining the dots
Scatter graph points should not be connected one by one. If requested, draw one straight line of best fit through the overall pattern.
Forcing the line through the origin
The line should pass through the centre of the data pattern. It should only go through (0,0)(0,0)(0,0) if the trend genuinely supports that position.
Treating an estimate as exact
Values read from a line of best fit are estimates. Use “approximately” or the symbol ≈\approx≈ where appropriate.
Claiming causation
Positive or negative correlation does not prove that one variable causes changes in the other. State the association shown by the data without making an unsupported causal claim.
Extrapolating without recognising the risk
A line can be extended beyond the observed data, but such predictions are less reliable. The relationship may change outside the measured range.
Scatter graphs exam questions
Embedded exam-style question
A researcher records the number of hours that ten students revise and their percentage scores in a test. The scatter graph shows strong positive correlation. A line of best fit gives an estimated score of 72%72\%72% for 666 hours of revision.
Question
- Describe the relationship between revision time and test score. [1][1][1]
- Interpret this relationship in context. [1][1][1]
- The observed revision times range from 111 to 888 hours. Explain why using the line to predict the score for 121212 hours is less reliable. [1][1][1]
- Does the graph prove that extra revision causes a higher score? Explain your answer. [1][1][1]
Answer
- There is strong positive correlation.
- Students who revise for longer tend to achieve higher test scores.
- Twelve hours is outside the observed range, so this is extrapolation and the trend may not continue that far.
- No. Correlation shows an association but does not prove causation; other variables could affect test scores.
Try 4 more scatter graphs practice questions.
Four more practice questions
Question 1
A scatter graph compares the age of a laptop with its resale value. The points slope downwards closely from left to right. Describe the correlation and interpret it in context. [2][2][2]
Answer: There is strong negative correlation. As the age of a laptop increases, its resale value tends to decrease.
Question 2
A line of best fit is used to estimate a value within the range of the collected data. Name this type of estimation and explain why it is usually more reliable than estimating outside the range. [2][2][2]
Answer: It is interpolation. It is usually more reliable because it is supported by observed data on both sides of, or around, the estimated value.
Question 3
A point lies far away from the general pattern on a scatter graph. State the statistical term for this point and give one possible explanation. [2][2][2]
Answer: It is an outlier. It could be a measurement error, a recording error, an unusual case or a genuine atypical observation.
Question 4
A student draws a line of best fit by joining the first point to the last point. Explain why this method may be incorrect. [1][1][1]
Answer: A line of best fit should represent the overall trend and pass through the middle of the main cluster, with a reasonable balance of points on each side. The first and last points may not represent that trend.
How to revise scatter graphs effectively
Practise the separate skills before combining them in longer questions:
- identify positive, negative and no correlation
- describe relationships using the variables and units
- identify outliers
- draw sensible lines of best fit
- make estimates in either direction
- distinguish interpolation from extrapolation
- explain why correlation does not establish causation
When marking your work, allow for sensible variation in graphical estimates. For written questions, compare your wording with the mark scheme and check that you referred to the context rather than giving only a general definition.
Scatter graphs can appear within broader statistics questions, so use GCSE past papers to practise recognising the required method without being told the topic. You can also review the GCSE Maths topics by exam board and tier to place this skill within your wider revision plan.
Conclusion
For scatter graphs GCSE Maths questions, remember that each point represents paired numerical data. Describe the direction and strength of correlation, interpret it in context, draw a balanced line of best fit and treat values read from that line as estimates.
Interpolation is normally more reliable than extrapolation, an outlier should not control the line of best fit, and correlation does not prove causation. Use the MathsGenie scatter graphs practice questions, mark schemes and past papers to apply these ideas under exam conditions.



