d effect size calculator

Cohen’s d Calculator (Independent Groups)

Use this tool to calculate Cohen’s d, a standardized mean difference between two groups. Enter each group’s mean, standard deviation, and sample size.

Assumptions: independent groups, approximately normal distributions, and pooled SD method.

What is d effect size?

In statistics, Cohen’s d is one of the most common ways to describe the size of a difference between two means. Unlike a p-value, which only tells you whether an effect is statistically detectable, d tells you how big that effect is in standardized units.

Put simply: Cohen’s d answers the question, “How far apart are the two groups, relative to their typical variation?”

Formula used in this calculator

This calculator uses the pooled-standard-deviation version for independent samples:

spooled = sqrt( ((n1 - 1)SD12 + (n2 - 1)SD22) / (n1 + n2 - 2) )

d = (M1 - M2) / spooled
  • M1, M2 are group means.
  • SD1, SD2 are standard deviations.
  • n1, n2 are sample sizes.

How to interpret Cohen’s d

A commonly used guideline from Cohen is:

  • 0.2 = small effect
  • 0.5 = medium effect
  • 0.8 = large effect

These are rough heuristics, not hard rules. In some fields (like medicine), even a small effect can be important. In others (like education or psychology), practical relevance depends on context, cost, and feasibility.

Step-by-step: using the calculator

  1. Enter the two group means.
  2. Enter each group’s standard deviation.
  3. Enter sample sizes (must be at least 2 in each group).
  4. Click Calculate d.
  5. Review Cohen’s d, Hedges’ g, and interpretation output.

What else is reported?

Hedges’ g

For smaller samples, Cohen’s d can be slightly biased upward. The calculator also reports Hedges’ g, which applies a correction factor.

95% confidence interval for d

A confidence interval gives a plausible range for the true standardized effect size. Wider intervals indicate more uncertainty.

Common-language effect size

This is an intuitive estimate of the probability that a randomly selected person from Group 1 scores higher than a randomly selected person from Group 2 (under normality assumptions).

Common mistakes to avoid

  • Using d for paired/repeated-measures data without the correct paired formula.
  • Interpreting sign incorrectly: positive d means Group 1 mean > Group 2 mean.
  • Relying only on p-values and ignoring effect size magnitude.
  • Forgetting confidence intervals when reporting uncertainty.

When to use (and not use) this calculator

Good fit

  • Two independent groups
  • Continuous outcome
  • Need a standardized difference for reporting or meta-analysis

Not ideal

  • Paired samples (use paired-samples effect size)
  • Strongly non-normal outcomes with severe outliers
  • Binary outcomes (consider odds ratio, risk difference, etc.)
Reporting tip: A clean sentence might look like: “The intervention group scored higher than control, d = 0.62, 95% CI [0.21, 1.03], indicating a medium effect.”

Quick FAQ

Is a negative d bad?

No. Negative only indicates direction (Group 1 lower than Group 2).

Can I compare d across studies?

Yes, that is one reason d is widely used in meta-analysis, as long as constructs and measures are reasonably comparable.

Should I always report both d and p?

In most research contexts, yes. p-values address statistical evidence; effect sizes address practical magnitude.

🔗 Related Calculators