Causality

A finding that two things tend to change together (rise or fall in step).

When one thing makes another happen. It is a stronger claim than correlation.

Two things can move together without one causing the other, so causal claims need more than a correlation.

A hidden factor that causes both of the things measured. Hot weather raises ice-cream sales and drownings.

The direction is the opposite of what was assumed: B leads to A, not A to B.

An experiment: it changes one thing on purpose and keeps everything else the same.

It cannot rule out third variables or reverse causation, or control other factors.

Towns with more churches have more crime. Both come from a bigger population, a third variable.

It is one of four concepts in the 15-mark question on a study you are given, and it can be asked about your class practical.

Spotting a real link that is worth testing properly with an experiment.

The one thing the researcher changes on purpose to see its effect.

The thing the researcher measures to see the effect of the IV.

It changes the IV and keeps everything else the same, so a change in the DV can only come from the IV.

A group that does not get the IV, used as a comparison.

Assigning participants to groups by chance, so the groups start out similar.

It spreads third variables evenly across groups, so they can't explain the result.

Other factors that could affect the DV; they must be controlled (kept equal).

Confidence that the IV, and nothing else, caused the change in the DV.

The experiment controls other factors and uses random allocation, ruling out third variables and reverse causation.

How far a finding holds outside the study: for other people, places and real life.

It gives the control group the same expectation of being treated, so expectation cannot explain a difference.

A control group that gets the treatment later. It is used when no placebo exists, as with therapy.

The difference is unlikely to be due to chance alone. It does not mean large or important.

One where two things each cause the other, not just one direction.

A cycle where each thing makes the other stronger, repeating over time.

Stress and sleep: stress makes sleep worse, and poor sleep makes stress worse.

They show a simple, one-way 'A causes B' is often too simple for real behaviour.

Low mood leads to withdrawing, and withdrawing makes mood lower. Each feeds the other.

How each direction could work in the study, and that both may be true in a loop.

You can break any part of the loop (e.g. improve sleep) to slow the whole cycle.

Bidirectional: the two things cause each other. Third variable: a hidden factor causes both.

Not being able to tell, from a study, which of two related things affects which.

Not a line, a loop — the arrow points both ways.

What one factor does depends on the other. Poor sleep may matter most for students who are also anxious.

Measure the same people more than once and see which change comes before the other.