A complete meta-analysis, no R required
Covexe runs the analysis most researchers otherwise need R, Stata, or a paid desktop package for. Upload your study data as CSV or Excel, choose an effect measure (odds ratio, risk ratio, hazard ratio, mean difference, or standardized mean difference) and a fixed-effect or random-effects model, and you get the pooled estimate, a publication-ready forest plot, and the full heterogeneity picture: Cochran's Q, I², and tau² with DerSimonian-Laird or REML estimation.
Beyond the pooled estimate
- Subgroup analysis and meta-regression
- Leave-one-out sensitivity analysis
- Funnel plot with Egger's and Begg's tests for publication bias
- Network meta-analysis when you have more than two treatments
- SVG and PNG figure export, CSV and Excel result export
Numbers you can defend
Every method is validated against R (metafor for pairwise meta-analysis, netmeta for networks), and every result shows the formula and the citation behind it. You can download R replication code for the analysis, so a reviewer or supervisor can reproduce your numbers independently.
The steps before the analysis live here too
Most free meta-analysis calculators start and end at the forest plot. Covexe covers the review around it: question framing and search strategy, abstract screening with a second reviewer and Cohen's kappa, AI-assisted data extraction from PDFs, risk of bias assessment (RoB 2, ROBINS-I, NOS, JBI, AXIS), GRADE certainty ratings, and the PRISMA 2020 flow diagram that documents it all. One project, one place, free core tools.
How it compares
RevMan is free but built around the Cochrane workflow. Comprehensive Meta-Analysis costs 495 to 1395 USD and runs on the desktop. R is free and powerful but asks you to code. Covexe sits in the gap: point and click in the browser, statistics checked against R, and the systematic review workflow included rather than sold separately.