FREE TOOL
Enter each study's effect size and standard error to compute Cochran's Q, I², and τ² — the same formulas EvidenceFlow's meta-analysis engine uses.
Cochran's Q
1.79
I²
0.0%
τ²
0.0000
If your effect measure is a ratio (OR, RR, HR), enter effect sizes and standard errors on the log scale — the same convention Cochrane reviews use.
Cochran's Q is computed as the inverse-variance-weighted sum of squared deviations from the fixed-effect pooled estimate: Q = Σ wi(θi − θ̂)², where wi = 1/SEi².
I² expresses the proportion of total variation due to genuine heterogeneity rather than chance: I² = max(0, (Q − (k−1)) / Q) × 100%, where k is the number of studies. Higgins & Thompson's convention treats I² below 25% as low, 25-75% as moderate, and above 75% as high heterogeneity.
τ² (tau-squared) is the DerSimonian-Laird estimate of between-study variance, used to derive random-effects weights in a full meta-analysis.
FAQ
I² is the percentage of total variability across studies that is due to genuine heterogeneity rather than sampling error — 0% means all variation is chance, higher values mean studies disagree beyond what chance would predict.
By the commonly used Higgins & Thompson thresholds: under 25% is low, 25-75% is moderate, and over 75% is considered high heterogeneity — though these are guides, not strict cutoffs.
For ratio measures like odds ratio, risk ratio, or hazard ratio, enter the effect size and standard error on the log scale, since pooling and heterogeneity statistics are only valid on that scale.
Q is a heterogeneity test statistic that depends on the number of studies; I² converts Q into a percentage that's comparable across meta-analyses of different sizes; τ² estimates the actual variance between true study effects, used to weight studies in a random-effects model.
Need to pool more than heterogeneity — forest plots, subgroup analysis, Mantel-Haenszel and Peto pooling? EvidenceFlow runs the full meta-analysis from your extracted data.
Create free account