External variables and causality

IB Psychology· Research methodology· 11 min read

1. Core concept

External (extraneous and confounding) variables are factors other than the independent variable that could influence the outcome. If they are not controlled, they offer rival explanations, so a difference or correlation cannot safely be read as cause and effect. Experiments try to neutralise them through control, randomisation and standardisation; non-experimental studies cannot, which is why correlation does not imply causation. Drawing a causal conclusion means first ruling out the plausible external variables.

2. Key studies

IB Psychology answers must be supported by named studies. Learn these to cite as evidence:

📘 Definition

Stein & Friedrich (1972)

Aggressive television increased aggression only in already-aggressive children, and prosocial television helped mainly children from lower-SES backgrounds. — The children's prior aggression and background are external variables that moderate the effect — ignore them and you would wrongly read a simple 'TV causes behaviour' relationship.

3. Evaluation (AO3)

Strength — the key to causal claims — Identifying and controlling external variables is exactly what lets experiments claim cause where correlational studies cannot.

Limitation — you cannot control everything — In real settings countless variables operate, and controlling them creates artificiality — so some confounds always remain, and over-control costs ecological validity.

4. Scope

5. Common Pitfalls

Wrong move:

A strong or significant relationship between two variables shows one causes the other.

Why:

A frequently penalised misconception in IB Psychology exams.

Correct move:

An uncontrolled third variable can produce the association, or moderate it. Before claiming cause, ask what external variables were controlled or ruled out — correlation alone never settles it.

Going deeper

Types of research methodwhy experiments can, and non-experiments cannot, claim cause is RM_11.
Inferential statisticsa significant correlation still is not cause (RM_6).