Speca › Topics › GCSE Maths
Data representation and scatter graphs
AQA GCSE Maths
10 ready-made resources for teaching Data representation and scatter graphs, written for AQA GCSE Maths. Slides, worksheets, mark schemes, homework, an assessment and retrieval quizzes, with SEN-friendly lesson design as standard. Below is what the topic requires and where students usually go wrong, free to read whether or not you sign up.
Get these 10 files free
What is in this bundle
- Assessment mark schemeWord, editable
- End-of-topic assessmentWord, editable
- HomeworkWord, editable
- Retrieval gameHTML, editable
- Retrieval game answer sheetWord, editable
- Retrieval-practice quizzesWord, editable
- SlidesPowerPoint, editable
- Starter & exit ticketWord, editable
- Tiered worksheetWord, editable
- Worksheet mark schemeWord, editable
Stuck? Worked examples, step by step
The same methods the lessons teach, one step at a time. Cover the working, try it yourself, then check each step.
- Reading bar charts and pictograms
- Reading a pie chart
- Reading a frequency polygon
- Reading a time-series graph
- Drawing a line of best fit
- Strength of correlation, and causation
- Predicting beyond the data (extrapolation)
- Calculating frequency density
- Random sampling and bias
- Stratified sampling
Foundation
Reading bar charts and pictograms
Bar chart: read the height against the scale.
Pictogram: count the symbols, then multiply by the key.
Foundation
Reading a pie chart
60 students were asked their favourite subject. Maths has an angle of 90°.
fraction = angle ÷ 360
= 90 ÷ 360 = ¼
number of students = ¼ × 60 = 15
Foundation
Reading a frequency polygon
Each point is plotted at the class midpoint, at a height equal to the frequency.
The points are joined with straight lines: this shows the overall shape of the distribution.
Foundation
Reading a time-series graph
Monthly sales, in £000s.
Sales rose steadily from January, dipped slightly in March, then rose sharply into April.
The steepest section of the line shows the biggest change.
Foundation
Drawing a line of best fit
Hours revised vs. test score.
A line of best fit passes through the middle of the points, with roughly equal numbers above and below.
Use it to estimate a score for any number of hours within the data range (1 to 8 hours).
Higher
Strength of correlation, and causation
Strength: how close the points are to the line.
Close to the line = strong correlation. Spread out = weak correlation.
Causation: correlation shows a relationship, but doesn't prove one thing causes the other.
Example: ice cream sales and sunburn cases both rise in summer, they're correlated, but ice cream doesn't cause sunburn. Hot weather causes both.
Higher
Predicting beyond the data (extrapolation)
The revision data covers 1 to 8 hours.
Predicting a score for 25 hours means extending the line far beyond the data collected.
This is called extrapolation, and it's unreliable: the relationship might not continue in the same way outside the range that was actually measured.
Higher
Calculating frequency density
68 swimmers' times (seconds).
Class 30-40: frequency 24, width 10
24 ÷ 10 = 2.4
Class 40-70: frequency 27, width 30
27 ÷ 30 = 0.9
Taller bar = more densely packed data, not necessarily more people.
Higher
Random sampling and bias
Random sampling: every member of the population has an equal chance of being chosen, for example, giving each person a number and picking numbers using a random number generator.
A biased sample doesn't represent the population fairly, so conclusions drawn from it can be wrong.
A school wants to know students' favourite lunch option. A teacher only surveys students queuing in the canteen at 12:30pm. Why is this biased?
This only samples students who eat in the canteen at that exact time: it misses students with packed lunches, different lunch times, or who skip lunch, so the results don't represent the whole school.
Higher
Stratified sampling
A stratified sample keeps the same proportions as the whole population.
A school has 800 students: 350 in Year 10, 450 in Year 11.
A stratified sample of 80 students is needed.
Year 10: (350 ÷ 800) × 80 = 35
Year 11: (450 ÷ 800) × 80 = 45
Foundation tier
- Draw and interpret bar charts, pictograms, and pie charts.
- Draw and interpret frequency polygons.
- Draw and interpret time-series graphs.
- Draw and interpret scatter graphs, and identify positive, negative, or no correlation.
- Draw a line of best fit by eye and use it to make predictions within the range of the data.
Higher tier, in addition
- Interpret the strength of correlation shown by a scatter graph, and understand that correlation does not imply causation.
- Use a line of best fit to make predictions beyond the range of the data, and discuss how reliable this is.
- Draw and interpret a histogram with unequal class widths, using frequency density.
- Understand basic sampling methods, including random and stratified sampling, and identify potential sources of bias.
Where students go wrong
- Misreading a pie chart by not correctly converting an angle into a proportion or frequency.
- Assuming correlation shown on a scatter graph implies one variable causes the other.
- Treating a prediction made far beyond the data range (extrapolation) as equally reliable to one made within it.
- Plotting frequency instead of frequency density on a histogram when class widths are unequal.
- Using the wrong ratio when calculating a stratified sample, rather than applying the correct proportion to each stratum.
How it gets asked in the exam
"Draw a line of best fit", "Describe the correlation shown", "Use your line of best fit to estimate...", "Calculate the frequency density", "Work out how many should be sampled from each...".
Key vocabulary
Bar chart, pictogram, pie chart, frequency polygon, time-series graph, scatter graph, correlation, line of best fit, interpolation, extrapolation, histogram, frequency density, sampling, stratified sample, bias.
Assumed prior knowledge
- Averages and spread.
- Ratio and proportion, needed for stratified sampling calculations.
- Plotting and reading coordinates.
How Speca scaffolds this topic
- Provide a "chart chooser" decision flowchart: a short sequence of yes/no questions (for example "showing parts of a whole?" leads to a pie chart; "showing change over time?" leads to a time-series graph) so the right chart type is selected systematically rather than guessed.
- Use a colour-coded correlation reference strip (positive, negative, none), each paired with a small icon scatter-pattern, shown consistently whenever a scatter graph question appears.
- For histograms, provide a worked "frequency density = frequency ÷ class width" formula card with the class-width row and frequency row colour-linked directly to the bar being drawn, extending the colour-linking pattern already used for grouped-data bar charts in the averages and spread topic.
Every file, free to start
Three full bundles a month at no cost, no card needed. Speca can also write a resource for a topic we have not built yet, and mark your students' answers to this lesson's questions against its own mark scheme.
Start free