An association is a pattern in which two things change together, and a causal claim says one produces the other. The first can be seen in a table; the second needs more than a table, so wording matters.
This skill is used whenever an issue source says “X leads to Y”. It follows recognising an unrepresentative sample inside data interpretation and limitations. All data here are fictional.
How do you test a causal claim?
Work through these checks:
- Direction: could the cause and effect be the other way round?
- Third factors: is something else shaping both?
- Chance: could a small number of cases produce this pattern by luck?
- Timing: did the supposed cause come first?
- Mechanism: is there a believable route from cause to effect?
- Comparison: is there a fair comparison group?
The more checks are met, the stronger the wording can be.
Worked example (fictional)
A made-up study of 8 classes finds that classes with more library visits have higher average reading scores.
| Class | Library visits per student (term) | Average reading score |
|---|---|---|
| 1 | 2 | 54 |
| 2 | 4 | 58 |
| 3 | 6 | 63 |
| 4 | 9 | 70 |
(Four of the eight rows are shown.) The headline says: “Library visits raise reading scores.”
Direction: better readers may enjoy the library more, so scores could drive visits.
Third factors: classes with more supportive home reading habits might do both.
Chance and size: 8 classes is a small number.
Comparison: no class was given extra visits for the study, so the data compare existing groups.
Fair interpretation: library visits and reading scores are positively associated in these 8 classes. The data do not show that visits cause higher scores.
Fair sentence: “Classes that visited the library more had higher average scores; this fits the idea that reading practice helps, but the study cannot rule out other explanations.”
What is the common mistake?
Mistaken answer: “The table proves that more library visits cause better reading.”
The student saw a pattern and treated it as a mechanism. The correction is to name at least one alternative explanation, such as reverse direction or a third factor, and to replace “proves” and “causes” with wording the evidence supports.
Check yourself
All data are fictional.
1. In a fictional town, ice cream sales and sunburn cases both rise in July. Name a likely third factor.
Show answer
Hot, sunny weather increases both. Ice cream need not cause sunburn.
2. Students who sleep longer score higher in a fictional survey. Give one way the cause could run the other way.
Show answer
Students who feel confident about their work may worry less and sleep better, so higher achievement could lead to better sleep.
3. Choose the better wording: (a) “Cycling to school makes students fitter.” (b) “In this sample, students who cycled to school scored higher on a fitness test.”
Show answer
(b). It describes what was measured. Version (a) claims cause, which a one-off comparison cannot show.
Where does this lead next?
Even careful claims depend on complete data, so the final lesson is communicating missing data without inventing values. Then try the integrated practice set.
Careful wording is a skill that improves with feedback, and that is where online one-to-one Global Perspectives tuition can help.