Effect Size
Effect size quantifies the strength of an effect or the magnitude of a difference between groups. Common measures include Cohen's d (difference between means), correlation coefficient r, or odds ratios. Effect size is independent of sample size, unlike p-values which depend on both effect size and sample size. A result can be statistically significant (small p-value) with negligible effect size.
Real World
A UK government study found a statistically significant but tiny effect of class size on GCSE grades (p < 0.001) with a Cohen's d of 0.05 — a practically negligible difference despite the large sample of 200,000 pupils.
Exam Focus
Distinguish statistical significance (p-value) from practical importance (effect size) — examiners frequently test this distinction in 'evaluate' questions.
How well did you know this?