Search
Subjects and activities

Class 10 · Science and environment

Together is not the same as because

In one town, months with more ice-cream sales also have more heatstroke cases. Does ice cream cause heatstroke?

What you will learn: Tell correlation from causation and name a third factor.

Small test at the end: scroll down to the end of this page for five short questions that check what you have learnt. Tap an answer, then press Check answer. Go to the questions ↓

Jump to: Check yourself · For grown-ups · For teachers

Safety first

No special risks.

You will need

  • The data on this pageAll made up for this lesson.
  • Paper and pencil

Time: About 40 minutes.

The story

Tanmay reads a news line: "Months with more ice-cream sales also have more heatstroke cases." He pushes his kulfi away. "Ice cream causes heatstroke!" Dhruv raises an eyebrow. "Or is something else going on?" Does going up together mean one causes the other?

The idea

Correlation means two things change together: when one is high, the other tends to be high too (or low).

Correlation does not prove causation. A third factor may drive both, the cause may run the other way, or it may be chance.

Here hot weather is a likely third factor: heat makes people buy ice cream, and heat causes heatstroke.

To test whether one thing causes another, change only that thing in a fair test and compare, or compare groups that are alike in every other way.

A table of made-up data for five months. March: 32 °C, 200 ice creams, 4 heatstroke cases. April: 36 °C, 350 and 7. May: 41 °C, 500 and 10. June: 39 °C, 450 and 9. July: 34 °C, 300 and 6. All three rise and fall together.Five months in one town (made-up data)MonthMax tempIce creams soldHeatstroke casesMar32 °C2004Apr36 °C3507May41 °C50010Jun39 °C4509Jul34 °C3006
A table of made-up data for five months. March: 32 °C, 200 ice creams, 4 heatstroke cases. April: 36 °C, 350 and 7. May: 41 °C, 500 and 10. June: 39 °C, 450 and 9. July: 34 °C, 300 and 6. All three rise and fall together.

Worked example

Analysing the ice-cream claim

  1. Claim: "Ice cream causes heatstroke."
  2. Pattern: sales and cases rise and fall together, highest in May.
  3. Third factor: temperature is also highest in May. Heat could cause both.
  4. Test idea: compare days with the same temperature but different ice-cream sales. That helps investigate the claim, but it does not settle it: other things can still differ between those days (rain, school holidays, how many people were outside), and a few days is a small sample. Name what else could differ before you decide.

The data show correlation. Hot weather explains both, so the claim is not supported.

Try it

  1. State the claim.

    Write: "A causes B."

  2. Describe the pattern.

    Do they rise and fall together? When are both highest?

  3. Look for a third factor.

    What else changes at the same time?

  4. Consider the other direction.

    Could B cause A instead?

  5. Design a test.

    How could you change only A, or compare alike groups, to check?

What you make (paper or offline)

Your claim analysis

A table showing temperature, ice-cream sales and heatstroke cases rising and falling together.Five months in one town (made-up data)MonthMax tempIce creams soldHeatstroke casesMar32 °C2004Apr36 °C3507May41 °C50010Jun39 °C4509Jul34 °C3006
A table showing temperature, ice-cream sales and heatstroke cases rising and falling together.

The claim and the pattern in the data.

A possible third factor and the other direction.

A test design that could tell them apart.

Make it on paper. Keep it with you. Nothing to upload.

Check yourself

Five questions about what this page teaches. Choose an answer and press Check answer, or open "See the answer and why". Nothing is saved or sent: your choices stay on this screen and disappear when you leave or reload the page.

  1. Question 1 of 5Ice-cream sales and heatstroke cases rise together. What is the best explanation?

    See the answer and why
    • A. Hot weather increases both Correct. Heat makes people buy ice cream and also causes heatstroke.
    • B. It proves the two cannot be linked at all Not this one. They are linked, through the weather.
    • C. Ice cream causes heatstroke Not this one. Rising together is not proof that one causes the other.
    • D. Heatstroke makes people buy ice cream Not this one. That reverses the direction, and heat explains both more simply.
  2. Question 2 of 5What does correlation mean?

    See the answer and why
    • A. A fair test with one change Not this one. That is a way to test causation.
    • B. Two things change together Correct. It describes a pattern, not a cause.
    • C. One thing causes another Not this one. That is causation, which correlation alone does not prove.
    • D. A kind of bar chart Not this one. Correlation is a relationship, not a chart.
  3. Question 3 of 5Students who sleep more get better marks. Which is a possible third factor?

    See the answer and why
    • A. Good marks making students sleep more Not this one. That is the cause running the other way, not a third factor.
    • B. A calm home routine that helps both sleep and study Correct. It could drive both sleep and marks.
    • C. The colour of their pillows Not this one. There is no reason it would affect both.
    • D. There cannot be any third factor Not this one. There often can be. That is why we look for one.
  4. Question 4 of 5How could you test whether one thing really causes another?

    See the answer and why
    • A. Ask people what they believe Not this one. Beliefs are not a test.
    • B. Collect more of the same kind of data Not this one. More of the same still shows only that they go together.
    • C. Change only that one thing in a fair test and compare Correct. If only that changed, a difference points to it as the cause.
    • D. Draw the graph bigger Not this one. A bigger graph shows the same correlation.
  5. Question 5 of 5In the table, what happened in May, the hottest month?

    See the answer and why
    • A. Ice-cream sales were high but heatstroke cases were low Not this one. Both were at their highest.
    • B. Heatstroke cases were lowest Not this one. Cases were highest in May, at 10.
    • C. Ice-cream sales and heatstroke cases were both highest Correct. Both peaked in the hottest month.
    • D. Ice-cream sales were lowest Not this one. Sales were highest in May, at 500.

Think back

Have you heard a claim that mixed up "together" with "because"?

Why is it tempting to believe the first explanation?

Investigate further

  • Find a news headline that claims a cause. What third factor could explain it?
  • Make up two things that rise together for a silly reason.

For grown-upsTap a heading to open it. The safety notes are at the top of the page.

Before you start
  • Nothing needed beyond this page.
Questions that help
  • "What else changed at the same time?"
  • "Could it work the other way round?"
What's really going on

Confusing correlation with causation is one of the most common reasoning errors in news and advertising.

Scientists use controlled experiments or careful comparisons to test cause.

Other ways to do it
  • Use umbrellas sold and puddles on the road: rain causes both.

For teachersLearning aims, classroom ideas and what to look for.

Learning aims
  • Distinguish correlation from causation.
  • Identify confounding factors and design a test.
Subject links
  • Science: scientific method.
  • Mathematics: data and relationships.
  • Social science: reading the news.

These links are general. They have not been checked against a specific board syllabus.

In the classroom
  • Groups get correlation claims and must find a third factor and a test for each.
What to look for
  • Names a plausible third factor.
  • Test changes only one thing.
ProLearners Assist