Cohen d tamanho efeito

(μ₁−μ₂)/σ.
Created by
Renato Passos, Eng. de Software
Reviewed by
Renato Passos, Eng. de Software

Last updated: Apr 18, 2026

d
0,500

About this calculator

The Cohen d effect size calculator computes the standardized difference between two means, expressed as (μ₁−μ₂)/σ, where μ represents group means and σ is the standard deviation. It is used in statistical studies to assess the magnitude of an effect, regardless of sample size. It is particularly useful when comparing groups in experiments or observational studies.

The calculation considers the absolute difference between means and the variability within groups. Larger values indicate more significant differences between groups. For example, d = 0.2 is interpreted as a small effect, d = 0.5 as medium, and d = 0.8 as large, according to Cohen's conventional criteria. The formula assumes homogeneous variances; adjustments are needed for unequal variances.

It is applied in fields like psychology, medicine, and social sciences to quantify differences between treatments, methods, or conditions. It is common in power analysis to determine appropriate sample sizes. Caution includes verifying data normality and checking variance homogeneity before interpreting results.

When using the calculator, provide the means and combined or separate standard deviations of the groups. If variances differ, prefer alternative methods like Welch's test. Interpret results based on the study context, avoiding absolute generalizations. Negative values indicate direction, but the absolute value is used to assess magnitude.

Frequently asked questions

What is Cohen d effect size?

It is a metric that quantifies the standardized difference between two means, independent of sample size, allowing comparison of effects across different contexts.

How does the calculator work?

It applies the formula (μ₁−μ₂)/σ, where μ represents means and σ is the combined or average standard deviation of the groups, depending on settings.

When to use this calculator?

Use it to compare groups in experiments, hypothesis testing, or studies evaluating effect magnitude, such as clinical or educational research.

What if variances are unequal?

Use Welch's test instead of traditional Cohen d, or adjust the standard deviation to address heterogeneous variances.

How to interpret the calculated values?

Values near 0.2 indicate a small effect; 0.5, medium; and 0.8, large. Interpretation depends on the study's specific context.

Other Power calculators