Effect Size Calculator
Calculate Cohen’s d, Hedges’ g, or Glass’s Δ for your statistical analysis
Calculation Results
Comprehensive Guide: How to Calculate Effect Size in Statistical Analysis
Effect size is a quantitative measure of the magnitude of the experimental effect, representing the standardized difference between two means. Unlike statistical significance (p-values), effect size provides information about the practical significance of research findings, making it essential for meta-analyses and evidence-based decision making.
Why Effect Size Matters
- Beyond statistical significance: A study can be statistically significant (p < 0.05) but have a trivial effect size
- Comparability: Standardized effect sizes allow comparison across studies with different measures
- Meta-analysis: Essential for combining results from multiple studies
- Practical significance: Helps determine if the effect is meaningful in real-world terms
Common Effect Size Measures
| Measure | When to Use | Interpretation Guidelines | Formula |
|---|---|---|---|
| Cohen’s d | Comparing two means (t-tests, ANOVA) | Small: 0.2, Medium: 0.5, Large: 0.8 | (M₁ – M₂) / SDpooled |
| Hedges’ g | Small sample sizes (n < 20) | Same as Cohen’s d with correction | Cohen’s d × (1 – 3/(4df – 1)) |
| Glass’s Δ | When control group SD is more representative | Same as Cohen’s d interpretation | (M₁ – M₂) / SDcontrol |
| Pearson r | Correlational studies | Small: 0.1, Medium: 0.3, Large: 0.5 | Covariance / (SDx × SDy) |
Step-by-Step Calculation Process
-
Gather your data:
- Group 1 mean (M₁) and standard deviation (SD₁)
- Group 2 mean (M₂) and standard deviation (SD₂)
- Sample sizes for both groups (n₁, n₂)
-
Calculate the difference between means:
Numerator = M₁ – M₂
-
Determine the denominator based on effect size type:
- Cohen’s d: Pooled standard deviation (weighted average of both SDs)
- Hedges’ g: Same as Cohen’s d with small sample correction
- Glass’s Δ: Uses only the control group’s SD
-
Compute the effect size:
Divide the mean difference by your chosen denominator
-
Interpret the result:
Compare against established benchmarks (Cohen’s conventions)
Pooled Standard Deviation Calculation
The pooled standard deviation (used in Cohen’s d) accounts for both groups’ variability and sample sizes:
SDpooled = √[( (n₁ – 1)×SD₁² + (n₂ – 1)×SD₂² ) / (n₁ + n₂ – 2)]
Small Sample Correction (Hedges’ g)
For samples under 20, Hedges’ g applies this correction factor:
Correction = 1 – [3 / (4×(n₁ + n₂ – 2) – 1)]
Then multiply Cohen’s d by this correction factor
Effect Size Interpretation Guidelines
| Effect Size Value | Cohen’s Interpretation | Percentage Overlap | Real-World Example |
|---|---|---|---|
| 0.01 | Very small | ~99% | Height difference between 15- and 16-year-olds |
| 0.20 | Small | ~85% | Difference between male and female height |
| 0.50 | Medium | ~67% | Effect of tutoring on exam scores |
| 0.80 | Large | ~53% | Difference between high school and college reading levels |
| 1.20 | Very large | ~40% | Height difference between 13- and 18-year-olds |
| 2.00 | Huge | ~24% | Difference between children and adults in height |
Common Mistakes to Avoid
- Ignoring directionality: Effect size can be positive or negative – the sign indicates direction
- Using wrong denominator: Glass’s Δ uses control SD, Cohen’s d uses pooled SD
- Neglecting sample size: Small samples require Hedges’ g correction
- Misinterpreting benchmarks: Cohen’s guidelines (0.2, 0.5, 0.8) are general – field-specific standards may differ
- Confusing with p-values: A significant p-value doesn’t always mean large effect size
Advanced Considerations
For more complex designs:
- Partial eta squared (η²): For ANOVA designs with multiple factors
- Odds ratio: For binary outcomes in medical research
- Cramer’s V: For chi-square tests with categorical data
- Mahalanobis distance: For multivariate effect sizes
Practical Applications
Effect sizes are crucial in:
-
Education research:
Comparing teaching methods (e.g., traditional vs. flipped classroom)
-
Medical studies:
Assessing treatment efficacy (e.g., drug vs. placebo)
-
Psychology:
Evaluating therapy outcomes (e.g., CBT vs. control for anxiety)
-
Business:
Marketing A/B tests (e.g., conversion rate differences)
-
Policy evaluation:
Assessing social program impacts (e.g., job training effectiveness)
Software Tools for Calculation
While our calculator handles basic effect sizes, these tools offer advanced options:
- R packages:
compute.es,effsize,MBESS - Python libraries:
pingouin,scipy.stats - SPSS: Built-in effect size calculations in ANOVA and t-test outputs
- JASP: Free open-source alternative with effect size reporting
- G*Power: Power analysis software with effect size conversion
Reporting Effect Sizes
Best practices for academic and professional reporting:
- Always report the effect size with its confidence interval
- Specify which type of effect size you’re reporting (e.g., “Cohen’s d = 0.45”)
- Include the direction of the effect (positive/negative)
- Provide context for interpretation (compare to similar studies)
- Report alongside statistical significance, not instead of it
Frequently Asked Questions
What’s the difference between statistical significance and effect size?
Statistical significance (p-value) tells you whether an effect exists in your sample, while effect size tells you how large that effect is. A study can be statistically significant with a tiny effect size (especially with large samples), or non-significant with a large effect size (with small samples).
When should I use Hedges’ g instead of Cohen’s d?
Use Hedges’ g when your sample sizes are small (typically under 20 per group). The correction factor accounts for bias in the estimation of the population effect size that occurs with small samples. For larger samples, Cohen’s d and Hedges’ g converge to similar values.
How do I calculate effect size for more than two groups?
For designs with three or more groups (one-way ANOVA), you would typically use:
- Eta squared (η²): Proportion of variance explained by the group differences
- Partial eta squared (ηₚ²): Proportion of variance explained after controlling for other factors
- Omega squared (ω²): Less biased estimate of variance explained in the population
These can be calculated from your ANOVA output tables.
Can effect sizes be negative?
Yes, effect sizes can be negative, which simply indicates the direction of the difference. For example, if Group 1 has a lower mean than Group 2, Cohen’s d will be negative. The absolute value indicates the magnitude regardless of direction.
How do I combine effect sizes in a meta-analysis?
Meta-analysis combines effect sizes using these key steps:
- Convert all effect sizes to a common metric (often Hedges’ g)
- Calculate weights for each study (typically inverse variance weights)
- Compute a weighted average effect size
- Assess heterogeneity (e.g., using I² statistic)
- Consider moderator analysis if substantial heterogeneity exists
Software like Comprehensive Meta-Analysis (CMA) or R packages (metafor, meta) can perform these calculations.
Authoritative Resources
For further reading on effect size calculation and interpretation: