← Math & statistics

Statistics Interpretation

Effect Size Calculator

Other languages 日本語 | English

Calculate Cohen's d, Hedges' g, or eta-squared from summary inputs, then read a cautious magnitude guide beside the effect value. Use it to complement p-values, not replace them.

This page is for effect-size interpretation after you have a design or result in mind. Use t-test, ANOVA, or correlation for inferential analysis, and use this page when you need the standardized magnitude view next to those results.

How to use

  1. Choose the metric first: Cohen's d, Hedges' g, or eta-squared.
  2. Enter two-group summaries for d or g, or choose ANOVA sums or group summaries for eta-squared.
  3. Read the effect value and the rough magnitude guide together with the context of your field and decision.

Wave 3 statistics expansion

Three effect-size workflows from summary inputs

Use Cohen's d for a standardized mean difference between two groups when you want a quick practical-size view next to a t-test result.

Inputs

Use this when you have two independent groups and want a standardized mean difference beside a t-test result.

What the page means by effect size

Effect size vs p-value

A p-value tells you how compatible the observed data are with a null model. Effect size tells you how large the observed pattern is on a standardized or variance-share scale. A very small effect can still have a low p-value in a large sample, and a practically important effect can have an uncertain p-value in a small sample.

Use this page when you need the magnitude view next to a test result, not as a replacement for study design checks, confidence intervals, or domain-specific interpretation.

Where this fits in a statistics workflow

FAQ

Why calculate effect size if I already have a p-value?

A p-value tells you how compatible the data are with a null hypothesis. Effect size tells you how large the observed difference or explained variance is on a practical scale. They answer different questions and should be read together.

What is the difference between Cohen's d and Hedges' g?

Both summarize standardized mean differences. Hedges' g applies a small-sample correction to Cohen's d, so it is often preferred when sample sizes are modest.

Should I treat small, medium, and large as fixed truth?

No. Those labels are rough conventions, not universal rules. Practical importance depends on the field, outcome, cost of change, and decision context.

When should I use eta-squared?

Use eta-squared as a simple one-way ANOVA effect-size summary when you want the share of total sample variance associated with group membership. It complements the ANOVA F test but does not replace the need to inspect design assumptions or follow-up comparisons.

Does the share URL include my numeric inputs?

No. The share URL stores only lightweight settings such as the active effect-size mode and eta-squared input source. Entered numeric values stay in your browser.

Comments (optional)

To reduce load, comments are fetched only when needed.