Mathematics · Grade 8 · Patterns in Data · Term 3

Correlation vs. Causation

Understanding that correlation does not imply causation.

• Ontario Curriculum Expectations
• 8.SP.A.1

About This Topic

Correlation describes a relationship where two variables tend to change together, shown through scatter plots with positive, negative, or no patterns. Causation means one variable directly influences the other. Grade 8 students learn that correlation does not prove causation, as seen in examples like higher ice cream sales and shark attacks both increasing in summer due to warmer weather, a confounding variable.

This topic aligns with Ontario's Grade 8 Mathematics curriculum in the Patterns in Data strand, where students construct scatter plots, interpret lines of best fit, and analyze bivariate data. It sharpens skills in questioning data claims from news reports or studies, fostering statistical literacy essential for informed decision-making.

Active learning benefits this topic because students actively sort examples, debate scenarios, and create their own datasets. These hands-on tasks reveal lurking variables and spurious links, making abstract distinctions concrete and memorable while building confidence in data analysis.

Key Questions

1. Explain why correlation between two variables does not necessarily mean that one causes the other.
2. Differentiate between situations that show correlation and those that imply causation.
3. Analyze real-world examples to identify instances of correlation without causation.

Learning Objectives

• Analyze scatter plots to identify patterns of correlation between two variables.
• Explain the concept of a confounding variable and its role in spurious correlations.
• Differentiate between correlation and causation using real-world examples.
• Evaluate statistical claims in media reports for potential correlation without causation.
• Create a scenario demonstrating correlation without causation, identifying the confounding variable.

Before You Start

Constructing and Interpreting Scatter Plots

Why: Students need to be able to visually represent bivariate data and identify general trends before they can analyze correlations.

Identifying Patterns in Data

Why: Understanding how to recognize trends, clusters, and outliers in data sets is foundational for discussing relationships between variables.

Key Vocabulary

Correlation	A statistical relationship between two variables, indicating that they tend to change together. This can be positive, negative, or show no clear pattern.
Causation	A relationship where one event or variable is the direct result of another event or variable.
Confounding Variable	An unmeasured variable that influences both the supposed cause and the supposed effect in an observational study, leading to a false association.
Spurious Correlation	A relationship between two variables that appears to be causal but is actually due to chance or a confounding variable.

Watch Out for These Misconceptions

Common Misconception

If two variables increase together, one must cause the other.

What to Teach Instead

Stress that correlation shows association only; causation requires experiments or strong evidence. Group debates on examples like this help students uncover confounders, shifting from assumption to analysis.

Common Misconception

A strong line of best fit proves causation.

What to Teach Instead

Lines of best fit quantify correlation strength but ignore directionality or third variables. Hands-on plotting activities let students manipulate data, seeing how patterns mislead without context.

Common Misconception

No correlation means no causation.

What to Teach Instead

Absence of correlation in observed data does not rule out causation, especially with limited samples. Collaborative data hunts reveal this, as students explore subsets where links emerge.

Active Learning Ideas

Card Sort: Correlation or Causation?

Prepare cards with real-world scenario pairs, like homework hours and grades. In small groups, students sort cards into correlation, causation, or neither piles, then justify choices with evidence. Follow with whole-class share-out to discuss confounders.

35 min·Small Groups

Data Debate Stations

Set up four stations with datasets: ice cream/drownings, storks/births, exercise/grades, TV watching/violence. Pairs rotate, plot scatter plots, hypothesize causation, and note possible third variables. Groups present findings.

45 min·Pairs

Spurious Correlation Creator

Individually, students pick two unrelated variables, find or fabricate data showing correlation, and propose a fun third factor. Share in whole class gallery walk, voting on most convincing examples.

30 min·Individual

Real-World Data Hunt

Provide safe online datasets or printouts. Small groups identify correlations, test for causation using criteria like temporal order, and report with visuals. Debrief common pitfalls.

50 min·Small Groups

Real-World Connections

• Market researchers might observe a correlation between increased advertising spending and higher product sales. However, they must investigate if other factors, like seasonal demand or competitor actions (confounding variables), are the true drivers of sales, not just the advertising itself.
• Public health officials analyze data for correlations between lifestyle choices and disease rates. For example, a correlation between coffee consumption and lung cancer might be observed, but smoking (a confounding variable) is often the actual cause of both.

Assessment Ideas

Discussion Prompt

Present students with a scenario: 'A study shows that cities with more libraries have higher crime rates.' Ask: 'Does this mean libraries cause crime? What other factors might explain this relationship?' Guide them to identify potential confounding variables.

Quick Check

Provide students with three statements. Two should describe correlation without causation (e.g., 'Ice cream sales increase when drowning incidents increase') and one should describe causation. Ask students to circle the statement that shows causation and briefly explain why the other two are only correlations.

Exit Ticket

Ask students to write down one example of correlation without causation they have encountered or can imagine. They should also name the potential confounding variable that explains the observed relationship.

Frequently Asked Questions

examples of correlation without causation for grade 8 math

Classic examples include ice cream sales and drowning rates, both rising in summer due to heat, not causation. Stork populations and birth rates correlate in some regions from rural nesting habits. Nicknames of U.S. presidents and margarine consumption show spurious links. Use these in class to plot data and discuss third variables like time or location.

how to teach correlation vs causation in grade 8 Ontario math

Start with scatter plots from prior lessons, then introduce examples via card sorts. Have students apply criteria: temporal order, mechanism, and confounders. Connect to media literacy by analyzing ads or headlines claiming links. Assess through journals where they critique real datasets.

active learning strategies for correlation vs causation

Card sorts and data debates engage students kinesthetically, as they physically categorize and argue examples. Creating spurious datasets individually sparks creativity, while station rotations build collaboration. These methods outperform lectures by making students discover confounders themselves, deepening retention and critical thinking.

common student mistakes correlation causation grade 8

Students often assume perfect correlation equals causation or overlook sample size limits. They reverse cause-effect or ignore bidirectional influences. Address via peer review of plots and scenarios, where groups spot errors in each other's work, reinforcing criteria like experimental controls.