Cohen's d
Cohen's d = (mean₁ - mean₂) / s_pooled, where s_pooled = √[((n₁-1)s₁² + (n₂-1)s₂²) / (n₁+n₂-2)]. Interpretation: |d| < 0.2 negligible, 0.2 ≤ |d| < 0.5 small, 0.5 ≤ |d| < 0.8 medium, |d| ≥ 0.8 large. Cohen's d is independent of sample size and comparable across studies.
Real World
Educational psychologist John Hattie calculated Cohen's d for hundreds of classroom interventions; one-to-one tutoring produced d ≈ 0.79, near the 'large' threshold, suggesting a near-one-standard-deviation improvement in attainment.
Exam Focus
Use the pooled standard deviation in the denominator, not just one group's SD — a common error that loses accuracy marks.
How well did you know this?