Cause and Effect Reasoning

Many LNAT passages make claims about what causes what. "X leads to Y", "Y is the result of Z", "A is responsible for B". Causal reasoning is fundamental to argumentative writing — and it is also where many arguments go wrong. The LNAT frequently tests your ability to evaluate causal claims and identify errors in causal reasoning.

---

Correlation vs Causation

The single most important principle in evaluating causal reasoning is:

Correlation does not prove causation.

Two things occurring together (correlation) does not mean that one causes the other. This is the most commonly tested causal reasoning concept on the LNAT.

Example

"Countries with higher chocolate consumption have more Nobel Prize winners per capita."

This is a real statistical correlation. But it would be absurd to conclude that eating chocolate causes Nobel Prizes. Both variables may be correlated with a third factor — national wealth, which funds both luxury food consumption and research institutions.

---

Four Errors in Causal Reasoning

1. Post Hoc Fallacy (Post Hoc Ergo Propter Hoc)

"After this, therefore because of this."

Just because event B occurred after event A does not mean A caused B.

"The government introduced a new policing strategy in January. Crime fell in February. Therefore, the new strategy reduced crime."

Why it is flawed: The reduction in crime might have been due to seasonal patterns, weather, economic changes, or countless other factors. Temporal sequence alone does not establish causation.

2. Confounding Variables

A confounding variable (also called a "third variable") is an unrecognised factor that is actually responsible for the observed relationship between two variables.

"Children who eat breakfast every day perform better at school than those who do not. Breakfast therefore improves academic performance."

Confounding variable: Families where children eat breakfast regularly may also be more stable, better-resourced, and more attentive to their children's education. The better academic performance may be due to these family factors, not the breakfast itself.

Observed correlation	Possible confounding variable
Breakfast → better grades	Family stability and socioeconomic status
Ice cream sales → drowning rates	Hot weather (increases both)
Hospital visits → death rates	Severity of illness (causes both)

3. Reverse Causation

Sometimes the causal relationship is the opposite of what is claimed — the supposed effect is actually the cause.

"People who exercise regularly report higher levels of happiness. Exercise therefore makes people happy."

Reverse causation possibility: Perhaps happier people are more likely to exercise, rather than exercise making people happy. The causal arrow may point in the other direction.

4. Overlooking Alternative Causes

Even when a genuine causal relationship exists, the author may focus on one cause while ignoring other contributory factors.

"The decline in reading among young people is caused by social media."

Problem: Social media may be a factor, but other causes might include changes in school curricula, reduced funding for libraries, the availability of other entertainment, and changes in parenting styles. Identifying one cause and presenting it as the cause is an oversimplification.

---

How the LNAT Tests Causal Reasoning

Question Type 1: "The author's reasoning is flawed because..."

The correct answer often identifies a causal reasoning error — such as confusing correlation with causation or committing the post hoc fallacy.

Question Type 2: "Which of the following, if true, would most weaken the argument?"

The correct answer often provides an alternative explanation (confounding variable or reverse causation) for the observed correlation.