I² Heterogeneity Calculator
Cochran's Q, I², H², τ² and pooled estimates for a meta-analysis.
Enter your values to see the result.
τ² uses the DerSimonian–Laird estimator and the I² interval uses the test-based method of Higgins & Thompson (2002). The Cochrane Handbook bands (0–40% might not be important, 30–60% moderate, 50–90% substantial, 75–100% considerable) overlap on purpose: judge I² together with the direction and size of effects.
Processing and privacy
Processing varies by tool. Check the instructions for local processing, server requests, external services and any file or request limits before entering sensitive information.
Read the privacy policyHow this calculation works
Understand the method, inputs and limits behind your result.
- Formula
- Q = Σwᵢ(effectᵢ − pooled effect)². I² = max(0, (Q − df)/Q) × 100, where df = studies − 1. Between-study variance uses the selected estimator.
- Variables
- Enter: Study effects, Q and k
- One study per line: effect, standard error. Use log odds or log risk ratios for ratio measures. A study label before the numbers is ignored.
- Cochran's Q
- Number of studies
- Confidence level: 90%, 95%, 99%
- Units
- Use the units shown beside each input. Counts and ratios are dimensionless unless labeled otherwise.
- Example
- Six studies with Q = 6.39 on 5 degrees of freedom give I² = 21.7%, which might not be important heterogeneity.
- Assumptions
- τ² uses the DerSimonian–Laird estimator and the I² interval uses the test-based method of Higgins & Thompson (2002). The Cochrane Handbook bands (0–40% might not be important, 30–60% moderate, 50–90% substantial, 75–100% considerable) overlap on purpose: judge I² together with the direction and size of effects.
- Last reviewed
- Review date not yet recorded.
Methodology & References
The method depends on the selected model and input options. Review the assumptions and references above before interpreting the result.
- Paste one study per line as effect, standard error (or enter Q and the number of studies).
- Choose the confidence level.
- Read I², its interval and the pooled estimates.
Interpretation and limitations
- How do I interpret I²? The Cochrane Handbook gives overlapping bands: 0–40% might not be important, 30–60% moderate, 50–90% substantial and 75–100% considerable.
- Why show the I² confidence interval? With few studies I² is very imprecise. A wide interval warns against reading too much into one value.
References
Share this tool
Found it useful? Send it to a friend or teammate.
Rate this tool
0.0
0 ratings
- 5 stars 0
- 4 stars 0
- 3 stars 0
- 2 stars 0
- 1 star 0
Clear instructions
Find steps, examples and limitations below.
Use online
Open the tool in a supported web browser.
Free to use
No sign-up required. Tool-specific limits may apply.
How to use the I² Heterogeneity Calculator
Cochran's Q, I², H², τ² and pooled estimates for a meta-analysis.
- 1 Paste one study per line as effect, standard error (or enter Q and the number of studies).
- 2 Choose the confidence level.
- 3 Read I², its interval and the pooled estimates.
Example and practical tips
Six studies with Q = 6.39 on 5 degrees of freedom give I² = 21.7%, which might not be important heterogeneity.
Further reading
Frequently asked questions
How do I interpret I²?
The Cochrane Handbook gives overlapping bands: 0–40% might not be important, 30–60% moderate, 50–90% substantial and 75–100% considerable.
Why show the I² confidence interval?
With few studies I² is very imprecise. A wide interval warns against reading too much into one value.
Practical guides
-
How to Use I² Heterogeneity Calculator
Calculate Cochran’s Q, I² with its confidence interval, H², τ² and fixed and random-effects pooled estimates from study effects and standard errors.
Report an issue
Something broken or not quite right? Tell us and we will look into it.