Question 1

What is Cohen's d and how is it computed?

Accepted Answer

Cohen's d is the standardized difference between two means: d = (mean₂ − mean₁) / sₚ, where sₚ is the pooled standard deviation sₚ = sqrt(((n₁−1)s₁² + (n₂−1)s₂²) / (n₁ + n₂ − 2)).

Question 2

How are odds ratios and their confidence intervals computed?

Accepted Answer

We compute OR = (a×d)/(b×c). The 95% CI uses the log‑scale Wald method: ln(OR) ± 1.96 × sqrt(1/a + 1/b + 1/c + 1/d), then we exponentiate the limits.

Question 3

Does Cohen's d assume equal variances?

Accepted Answer

Yes. Using the pooled SD assumes equal population variances across groups. If variances differ markedly, consider alternatives like Glass's Δ (uses the control SD) or Welch‑based approaches.

Question 4

What is Hedges' g and when should I use it?

Accepted Answer

Hedges' g applies a small‑sample correction to Cohen's d: g = J·d with J ≈ 1 − 3/(4(n₁ + n₂ − 2) − 1). For small samples, g is slightly less biased than d.

Question 5

How do I interpret the magnitude of d?

Accepted Answer

Common guidelines are |d| ≈ 0.2 (small), 0.5 (medium), 0.8 (large). Treat these as rough conventions—interpretation depends on your domain and context.

Question 6

Are odds ratios the same as risk ratios?

Accepted Answer

No. They are similar when outcomes are rare, but can diverge for common outcomes. Risk ratio compares probabilities; odds ratio compares odds.

Effect Size Calculator

Group 1

Group 2

Frequently Asked Questions

Explore More Tools