/en-tutor-prep-s3-maths-scatter-graphs-correlation/
/en-tutor-prep-s3-maths-scatter-graphs-correlation/
General Inquiry: info@TutorZone.com.hk
Office Hours: Monday to Saturday - 9:00 AM to 6:00 PM
WhatsApp Monday to Sunday - 24 hours
Hotline: +852 6828 1809
Instant Support: +852 9061 3106

S3 Maths | Scatter Graphs and Correlation: Tutor Preparation Guide

S3 Maths | Scatter Graphs and Correlation: Tutor Preparation Guide

  • 2026-11-21

Students see a cloud of points and announce a relationship, yet cannot separate correlation from cause, and often run a line of best fit far past the data when a question asks for a prediction. This lesson uses six students study hours and marks to link the scatter graph, direction and strength, and the line of best fit, then draws the boundary between interpolation and extrapolation.

Objectives — Students can (1) distinguish a scatter graph from a line graph and read direction and strength from the spread of points; (2) explain that correlation is not causation with one counterexample; (3) use a given line of best fit for an interpolation and explain why extrapolating beyond the data is unreliable.

01 Prerequisites (3-minute check)

If any of these fails, fix it before the lesson starts.

Skill Quick check Pass mark
Coordinate plane What does the point (3, 5) mean A pair of values, 3 across and 5 up
Mean How is the mean of six numbers found Total divided by the count
Straight line What is the gradient of y equals 2x plus 1 2, since y rises 2 for each step of x

02 Lesson flow (60 minutes)

Time Stage Tutor does Student does
0-6 min Open with a picture Show study hours against marks for six students Write one guess about any relationship
6-20 min Two kinds of graph Compare when each graph is used Give one example of each and say why no joining
20-34 min Direction and strength Read the cloud of points Label three graphs by direction and strength
34-46 min Line of best fit Anchor the line at the mean point Predict the mark at 4 hours
46-55 min Interpolation and beyond Stretch the same line to 10 hours Reach 102 marks and say why it fails
55-60 min Exit check Ask whether correlation implies cause Offer one counterexample

03 Teaching script (read it out)

Scene 1: Two graphs, two questions

The difference between a scatter graph and a line graph is the question, not the shape. A line graph asks how one object changes over time, so the points are joined in order; a scatter graph asks whether two different variables move together, and each point is a separate individual with no order between them. Joining them invents a trend.

Ask directly in class: is there an order among these six students? There is not, so the six points are never joined. One sentence, and the habit is broken.

Individuals have no order, so no joining

Scene 2: Correlation means direction and strength

Correlation is not a yes or no question about a relationship. It reports direction and strength: points rising left to right give positive correlation, falling give negative, and a shapeless cloud gives no clear correlation. Strength is how tightly the points hug an imagined straight line.

Close with the data. For these six students the correlation coefficient is about 0.93, a strong positive correlation, and yet a student who studied three hours scored above one who studied five. Correlation is a trend, never a promise.

Direction first, then tightness

Scene 3: A line may be drawn, but predictions stop at the data

The line of best fit passes through the centre of the data, so it must pass through the mean point, 3.5 hours and 57.3 marks. The gradient here is about 6.97 marks per hour, so four hours predicts roughly 60.8 marks, which is interpolation and is trustworthy inside the data range.

Stretch the same equation to ten hours and it returns about 102 marks, more than the paper is worth. That is the crack in extrapolation: the data stops at six hours, and no straight line extends forever. State that the extrapolation is unreliable rather than quoting the number.

The line may travel, the conclusion may not

04 Question types and worked steps

Type Example Steps Answer
Read the cloud Three scatter graphs Check direction, then tightness Strong positive, weak negative, no clear correlation
Predict from a line An equation and a value of x Substitute and judge interpolation or extrapolation Inside the range trustworthy, outside flagged as unreliable
Find the mean point Six pairs of data Average x and y separately The line must pass through (3.5, 57.3)
Correlation or cause Ice cream sales and drownings both rise Name the shared third factor Hot weather drives both, so no causation
Sanity check (teach this): Sanity check: a predicted mark cannot exceed full marks or fall below zero, so 102 signals extrapolation beyond the data; direction and gradient must share a sign, since positive correlation with a falling line means the axes were read backwards; the line of best fit must pass through the mean point, and a hand-drawn line that misses it is wrong; a correlation coefficient lies between minus 1 and 1, so 1.4 is a calculation error; outside the data range only a trend may be claimed, never a certain value; no clear correlation does not mean the variables are unrelated, only that a straight line describes them poorly.

05 Independent practice (3 questions)

Question Answer What to watch
Name the graph One scatter and one line graph Check whether the question type is named, not the shape
Predict twice At 4 hours and at 10 hours Check whether interpolation and extrapolation are labelled differently
Find a counterexample A correlated but non-causal pair Check whether the shared factor is identified

06 Common mistakes and what to say

Mistake Student writes Say this
Joining scatter points Drawing a line through six students Individuals carry no order; joining invents a trend, and only one object over time is joined
Correlation as cause Writing that more study raises marks, full stop Third factors must be excluded, and the answer should say the data shows correlation and needs further study
Extrapolating blindly Quoting 102 marks at ten hours Beyond the data the prediction must be flagged, and the maximum mark shows why
Reading the gradient backwards Calling a negative correlation positive Check the quantity and unit on each axis before judging the slope

07 Homework (paper, section and question numbers)

08 Exit check (last 5 minutes)

Question Answer Decision
Scatter graphs No joining, with a reason Redraw if the points are joined
Correlation Direction and strength Repeat the reading if only “related” is said
Extrapolation Say it is unreliable Return to the data range if the number is quoted

09 Tutor FAQ

Must the coefficient be calculated

At junior level the graph is enough to judge direction and strength; students should still know the coefficient runs from minus 1 to 1 and that values near the ends mean points hug the line.

Must the line be drawn by hand

Most questions supply the equation and test substitution and judgement; when drawing is required, the line must pass through the mean point with points spread evenly on both sides.

Does zero correlation mean no link

It does not. Two variables may be linked by a curve that a straight line cannot describe, so the safer wording is “no clear linear correlation”.

Find a tutor — freeTutor Registration →

A TutorZone lesson-prep guide. Written by the TutorZone team; questions and lesson flow are original and are not taken from any school or publisher material.

Join as a tutor or coach — academic, music, sports, arts, STEM. Free to register. See how it works →