Effect size calculator
The p-value tells you whether there is a difference or a relationship; the effect size tells you whether it is large or small. APA style asks for it next to every test, and many thesis committees now require it. Choose what you have: the means and standard deviations of two groups, or your test result (t, Z, F or chi-square) as SPSS reports it.
Enter the mean, standard deviation (SD) and number of people in each group.
The calculation runs in your browser. We don’t store what you type.
How it’s calculated
Cohen’s d from means: d = (M₁ − M₂) / pooled SD, where the pooled SD weights each group’s variance by its degrees of freedom. The 95% interval uses Hedges and Olkin’s standard error.
Hedges’ g: d multiplied by 1 − 3 / (4(n₁ + n₂) − 9). It corrects d’s bias with small groups (under 20 per group).
From Student’s t: independent groups, d = t × √(1/n₁ + 1/n₂); paired samples, dz = t / √n, with n the number of pairs.
From Z (Mann-Whitney U, Wilcoxon): r = |Z| / √N, with N the total number of cases.
From F (ANOVA): η² = F × df₁ / (F × df₁ + df₂). With one factor it equals the table’s η²; with several it is SPSS’s partial η². It also gives ω², which is less biased, and Cohen’s f, which power calculators ask for.
From chi-square: Cramér’s V = √(χ² / (N × df*)), with df* = the smaller of rows and columns, minus 1. In 2 × 2 tables V equals phi.
The probability of superiority puts d into words: the chance that a person from group 1 scores higher than a person from group 2.
How to read the result
- Cohen’s d: 0.2 small, 0.5 medium and 0.8 large. The sign only tells you which group has the higher mean.
- r: 0.1 small, 0.3 medium and 0.5 large.
- Eta squared: 0.01 small, 0.06 medium and 0.14 large. Read it as the share of variation the groups explain.
- Cramér’s V: cutoffs depend on table size. With df* = 1 they are 0.10, 0.30 and 0.50; with df* = 2, 0.07, 0.21 and 0.35. The calculator uses the cutoffs for your table.
- These are Cohen’s general guidelines. Your field may have its own benchmarks: if your advisor gives you others, use them.
In the glossary: Effect size · Cohen’s d
Next stepThesis methodology checklistDo you have everything a committee expects? Review your whole methodology with the checklist.Step-by-step guideEffect size: what it is and how to report it in APA 7 (with examples)What effect size is, which measure goes with each test (Cohen’s d, r, eta², Cramér’s V), how to interpret it and how to report it in APA 7.Frequently asked questions
- Why report effect size if I already have the p-value?
- With large samples tiny differences come out significant; with small samples important differences may not. Effect size doesn’t depend on sample size and tells you how much the result matters.
- Cohen’s d or Hedges’ g?
- They are nearly identical with 20 or more per group. With small groups report g, which corrects d’s upward bias.
- Which effect size goes with each test?
- Student’s t: Cohen’s d. Mann-Whitney U and Wilcoxon: r. ANOVA: eta squared. Kruskal-Wallis: epsilon squared. Chi-square: Cramér’s V. Correlation: r itself is the effect size.
- Does SPSS compute it?
- Since version 27, SPSS shows Cohen’s d and Hedges’ g in the t-test and eta squared in ANOVA if you tick the option. For r from Z and Cramér’s V from an existing table, this calculator saves you the step.
Know what you want to ask?
Build the survey in SurvIQ, share it by link, WhatsApp or QR code, and watch results come in live. The Free plan is free forever.
Create a free survey