Effect Size Interpretation
Effect size quantifies the magnitude of an effect without depending on sample size. Cohen's d = (mean₁ - mean₂) / σ standardizes the difference. Interpretation: d ≈ 0.2 (small), d ≈ 0.5 (medium), d ≈ 0.8 (large).
Real World
A study on Spotify's algorithm found that personalised playlists increased listening time with d = 0.3 (small effect), meaning the improvement, while real, was modest in everyday terms.
Exam Focus
Quote the d threshold (0.2/0.5/0.8) explicitly when classifying effect size as small, medium, or large to secure interpretation marks.
How well did you know this?