Scatter Graphs and Correlation

Active learning helps Year 11 students grasp scatter graphs because plotting real data builds intuition for patterns that abstract explanations alone cannot. Moving from hand-drawn points to debated lines of best fit makes the abstract concept of correlation concrete and memorable.

• Year 11
• Mathematics
• 4 activities
• 30 min–45 min

Learning Objectives

1. Analyze bivariate data sets to identify and classify the type of correlation present in a scatter graph.
2. Evaluate the strength and direction of correlation from a scatter graph, distinguishing between strong, moderate, and weak relationships.
3. Construct a line of best fit on a scatter graph by eye, justifying its placement relative to the data points.
4. Predict future values or trends using a line of best fit, and critique the reliability of these predictions based on the data's spread and the prediction's distance from the plotted data.
5. Explain the difference between correlation and causation, providing a reasoned example to support the explanation.

Case Study Analysis

⏱ 35 min·Pairs

Data Hunt: Class Measurements

Students pair up to measure partners' heights and arm spans using tape measures. Pairs plot their data on shared graph paper, discuss correlation type, and draw a line of best fit. Groups compare graphs and predict values for new students.

Prepare & details

Differentiate between positive, negative, and no correlation in scatter graphs.

Facilitation Tip: During Data Hunt, circulate to ensure students measure and record data accurately before plotting to avoid compounding errors later.

Setup: Groups at tables with case materials

Materials: Case study packet (3-5 pages), Analysis framework worksheet, Presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Case Study Analysis

⏱ 30 min·Small Groups

Correlation Card Sort: Real Datasets

Prepare cards with scatter plots, descriptions, and correlation types. Small groups sort cards into positive, negative, or none categories, then justify choices. Extend by drawing lines of best fit on selected plots.

Prepare & details

Explain why correlation does not imply causation.

Facilitation Tip: For Correlation Card Sort, group students heterogeneously so stronger peers model reasoning while struggling students contribute observations.

Setup: Groups at tables with case materials

Materials: Case study packet (3-5 pages), Analysis framework worksheet, Presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Case Study Analysis

⏱ 45 min·Whole Class

Prediction Challenge: Sports Data

Provide data on athletes' training hours and race times. Whole class plots on a large board, votes on line of best fit, then predicts performance for hypothetical athletes. Discuss prediction reliability based on scatter strength.

Prepare & details

Predict future values using a line of best fit, assessing the reliability of the prediction.

Facilitation Tip: In Prediction Challenge, require students to justify their line of best fit in writing before sharing estimates to prevent guesswork.

Setup: Groups at tables with case materials

Materials: Case study packet (3-5 pages), Analysis framework worksheet, Presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Case Study Analysis

⏱ 40 min·Small Groups

Causation Debate: Mystery Pairs

Show paired variables like shoe size and IQ. Individuals plot sample data, note correlation, then debate in small groups if one causes the other. Reveal lurking variables to clarify concepts.

Prepare & details

Differentiate between positive, negative, and no correlation in scatter graphs.

Facilitation Tip: During Causation Debate, assign roles to quiet students so everyone participates in challenging flawed causal claims.

Setup: Groups at tables with case materials

Materials: Case study packet (3-5 pages), Analysis framework worksheet, Presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management

Teaching This Topic

Teach scatter graphs by moving from concrete to abstract: start with students’ own measurements, then introduce varied real datasets to prevent overgeneralization to familiar contexts. Avoid rushing to the line of best fit; let students experience the messiness of real data first. Research shows that peer discussion of outlier placement improves understanding of balance and fit more than teacher correction alone.

What to Expect

Successful learning looks like students confidently plotting points, distinguishing correlation types, and using lines of best fit for reasoned predictions. They should articulate when correlation does not imply causation and adjust their lines based on class feedback.

Watch Out for These Misconceptions

Common Misconception

During Causation Debate, watch for students equating any upward trend with direct causation.

What to Teach Instead

Use the debate’s scenario cards to pause and ask each pair to list at least one alternative explanation for the observed link. Have volunteers share these to challenge assumptions before voting on correlation or causation.

Common Misconception

During Data Hunt, watch for students forcing the line of best fit to pass through every point.

What to Teach Instead

After plotting, ask groups to overlay their lines on a projected graph and hold up colored cards: green if their line balances points above and below, red otherwise. Discuss how outliers skew attempts at exact fits.

Common Misconception

During Correlation Card Sort, watch for students labeling any visible pattern as strong correlation.

What to Teach Instead

Ask students to rank each card by strength before grouping, then justify their rankings using distance from the imagined line of best fit. Circulate and ask probing questions about the spread of points.

Assessment Ideas

Exit Ticket

After Data Hunt, provide a printed scatter graph with five points showing a positive correlation. Students must: 1. Describe the correlation in one sentence. 2. Draw a line of best fit. 3. Predict the dependent variable value for an independent variable within the data range. Collect and use responses to identify students who balance points versus those who force lines through outliers.

Discussion Prompt

After Causation Debate, present Scenario A (ice cream sales and temperature) and Scenario B (shark attacks and ice cream sales). Ask students to write a paragraph explaining which shows correlation and which might show causation, referencing third variables like beach crowds or tourist season.

Peer Assessment

During Correlation Card Sort, have students swap their sorted groups with another pair and challenge their reasoning. Students must mark any disagreements and present their revised groupings with justifications at the end of the activity.

Extensions & Scaffolding

• Challenge students to find a dataset online with a non-linear pattern and explain why a straight line of best fit is inappropriate.
• Scaffolding: Provide pre-plotted graphs with gridlines color-coded for positive, negative, and no correlation to focus students on interpretation rather than plotting.
• Deeper exploration: Ask students to collect their own paired data over a week (e.g., hours of sleep and mood ratings) and present their analysis including correlation type and a justified prediction.

Key Vocabulary

Bivariate Data	A data set consisting of two variables for each individual observation, used to investigate relationships.
Correlation	A statistical measure that describes the extent to which two variables change together. It can be positive, negative, or none.
Line of Best Fit	A straight line drawn on a scatter graph that best represents the trend in the data, minimizing the distance between the line and the data points.
Causation	The relationship where one event is the result of another event; correlation does not imply causation.