Effect Size Calculator (Cohen's d + Hedges' g)

P-values tell you whether an effect is likely real; effect sizes tell you whether it\'s big enough to care about. A drug that reduces blood pressure by 0.5 mmHg in a 50,000-person trial will produce p < 0.001 — but the effect is clinically meaningless. Conversely, a 30-person pilot showing a 10-point IQ gain might fail to reach significance because the trial is underpowered — but the effect size (d ≈ 0.7) signals the intervention deserves a properly powered replication. Modern statistical guidance (APA, Open Science Framework, AMA) requires effect size + confidence interval reporting in every quantitative paper, alongside or instead of p-values.

The Cohen\'s d family

Cohen\'s d is the workhorse: standardized mean difference using pooled SD. Hedges\' g corrects d for small-sample bias — d overestimates the population effect when n is small, and the correction factor (1 − 3 / (4N − 9)) shrinks the estimate toward zero. Glass\'s Δ uses only the control group\'s SD as the standardizer, appropriate when the treatment is suspected to inflate variability. For correlational designs, the equivalent is r (Pearson correlation) or r² (proportion of variance explained). All are interconvertible: r ≈ d / √(d² + 4) for equal-sized groups.

"The primary product of a research inquiry is one or more measures of effect size, not p-values." — Cohen J. (1990) "Things I have learned (so far)", American Psychologist 45(12).

Cohen\'s benchmarks — useful, but contextual

Cohen\'s original 1988 textbook proposed 0.2 (small), 0.5 (medium), 0.8 (large) as benchmarks. These have been widely adopted but were intended as rough heuristics, not laws. Field-specific norms matter: in medicine, d = 0.2 for a low-cost intervention can be enormous practical value; in education, classroom interventions rarely exceed d = 0.4 — Hattie\'s "Visible Learning" meta-analysis of 800+ effect sizes found median d ≈ 0.4 across all educational interventions. Always interpret effect size against the field\'s base rate, not just against Cohen\'s ladder.

ASEAN research applications

Singapore Ministry of Education evaluates intervention programs (reading, math, ICT) using Cohen\'s d for between-cohort comparison. NUS, NTU, and University of Malaya psychology + medical departments report effect size in standard format. Indonesia + Philippines public health research increasingly uses Hedges\' g for small-trial pilot evaluation. Reporting d or g alongside p-values is now standard in journals like Asian Journal of Psychology, Singapore Medical Journal, and Asian Pacific Journal of Public Health.