Effect Size Calculator

What Is Effect Size?

Effect size measures how large a difference or relationship is, not just whether it is statistically significant. While a p-value answers "is there an effect?", effect size answers "how big is it?" This matters for practical significance, power analysis, and meta-analysis.

Cohen's d (Two Groups)

$$d = \frac{M_1 - M_2}{SD_{pooled}}$$

Hedges' g applies a small-sample correction to reduce upward bias in $d$. Glass's delta uses only the control group standard deviation when variances differ.

From t-Tests

For independent samples: $d = t\sqrt{\frac{1}{n_1}+\frac{1}{n_2}}$. For paired samples: $d_z = \frac{t}{\sqrt{n}}$.

From ANOVA

$$\eta^2 = \frac{F \times df_{between}}{F \times df_{between} + df_{within}}$$

Omega-squared ($\omega^2$) gives a less biased population estimate. Cohen's $f = \sqrt{\eta^2/(1-\eta^2)}$.

Frequently Asked Questions

What is a good Cohen's d value?

Jacob Cohen suggested 0.2 (small), 0.5 (medium), and 0.8 (large). Context matters: in some fields, 0.2 can be meaningful while in others 0.8 is common.

When should I use Hedges' g instead of Cohen's d?

Use Hedges' g when sample sizes are small (roughly under 20 per group). The correction reduces upward bias in Cohen's d for small samples.

What is CLES?

The Common Language Effect Size (CLES) converts Cohen's d into a probability: the chance a random person from Group 1 scores higher than a random person from Group 2.

What is the difference between eta-squared and omega-squared?

Eta-squared describes variance explained in your sample but tends to overestimate the population effect. Omega-squared is less biased and preferred for inference.

Report

What Is Effect Size?

Cohen's d (Two Groups)

From t-Tests

From ANOVA

Frequently Asked Questions

Related Tools

Recent Visits