Build your meta-analysis forest plot in the browser: fixed and random effects, heterogeneity (I², Q, τ²), funnel plot and a high-resolution PNG, no watermark.
Enter your studies and download a publication-ready chart as a high-resolution PNG. Calculating and viewing it is free; we only ask for an email to download the file, no watermark.
Pooled effect under a fixed-effect or random-effects model (with automatic selection), heterogeneity via Cochran’s Q, I² and tau², and a funnel plot to inspect publication bias. Automatic picks fixed effects when I² is low and Q is not significant; otherwise random effects.
The funnel plot compares each study's effect (X axis) against its standard error (Y axis). Without publication bias, points scatter symmetrically inside the triangle. Marked asymmetry, with small high-effect studies on only one side, is the classic sign of bias.
The effect size and its 95% confidence interval, as reported in each original study (Cohen's d, r, mean difference, ln(OR)…). Sample size is optional: it's only used for the total N in the results text, not in the calculation.
Method source: DerSimonian, R., & Laird, N. (1986). Meta-analysis in clinical trials. Controlled Clinical Trials, 7(3), 177–188.