Proportion Meta-Analysis Calculator: Pooled Prevalence with Forest Plot
Doing a prevalence review or a single-arm meta-analysis? Enter the number of cases and the sample size for each study and get the pooled proportion with its 95% confidence and prediction intervals, heterogeneity, subgroup results and a forest plot ready for your paper. Choose the logit or Freeman–Tukey double arcsine transformation and REML or DerSimonian–Laird random effects.
| Study | Cases (events) | Sample size (n) | Subgroup (optional) |
|---|
- PriceFree, no login or sign-up needed
- Your dataProcessed in your browser, never uploaded
- Works onPhone, tablet or computer, any modern browser
- Made byDr. Muhammad Aammar Tufail, Codanics
- Published17 November 2026
- Last updated16 November 2026
📚 How to cite this tool
This tool is free. If it helped your thesis, paper, report or assignment, please cite it. It keeps the tools free and helps other students find them.
Tufail, M. A. (2026). Proportion meta-analysis calculator: pooled prevalence with forest plot [Online tool]. Codanics. Retrieved 16 November 2026, from https://codanics.com/tools/proportion-meta-analysis-calculator/
In-text citation: (Tufail, 2026) · Narrative: Tufail (2026)
The access date is today’s date, filled in automatically. Author: Muhammad Aammar Tufail. Published 17 November 2026.
How to use the proportion meta-analysis calculator
- Enter one row per study: the number of cases (people with the condition, events or responders) and the sample size. Add a subgroup name (for example region or method) if you want subgroup results.
- Or click Paste from Excel / Upload CSV / Excel and match the columns. Load example shows ten simulated prevalence studies in three regions.
- Keep the logit transformation and REML random effects unless your protocol says otherwise. Choose how to show the results: percent, proportion or per 1,000.
- Click Pool proportions to get the pooled prevalence, 95% CI and prediction interval, I², subgroup results and the forest plot.
- Download the forest plot as PNG or SVG, copy the results paragraph and download the study-level table as CSV.
What a proportion meta-analysis does
A meta-analysis of proportions pools a single rate from each study, such as the prevalence of a disease, the proportion of patients who respond to a treatment in single-arm studies, or the proportion of seeds that germinate. Proportions near 0% or 100% have skewed sampling distributions and variances that depend on the proportion itself, so they are usually transformed, pooled on the transformed scale and then converted back to a proportion.
Methods and formulas
- For the logit and raw scales, a study with 0 cases or with all participants affected gets 0.5 added to the cases and to the non-cases, as in
metafor::escalc. The Freeman–Tukey and arcsine transformations handle 0% and 100% without a correction. - The prediction interval is μ̂ ± 1.96√(τ² + SE²) on the transformed scale, back-transformed, which is what
predict()in metafor reports. - Study CIs in the table are exact (Clopper–Pearson); the forest plot shows the back-transformed CI of each study on the chosen scale, so the two can differ slightly.
- The subgroup test pools each subgroup with its own τ², then compares the subgroup estimates with a fixed-effect Q test (the RevMan and
metadefault).
Logit or Freeman–Tukey?
Both are widely used. The double arcsine was popular because it stabilises the variance and copes with 0% and 100%, but Schwarzer et al. (2019) showed that its back-transformation can produce misleading pooled values when study sizes vary. Many methodologists now prefer the logit transformation (or a generalised linear mixed model in R). If you use Freeman–Tukey, report the logit result as a sensitivity analysis.
Worked example and validation
The example has ten simulated studies of a prevalence in three regions (2,225 participants). With the logit transformation and REML, the pooled prevalence is 36.9% (95% CI 31.8–42.2%), with a 95% prediction interval of 22.7–53.7% and I² = 85%. The regions differ (Asia 43.3%, Europe 29.8%, Africa 35.7%; Q = 7.28, df = 2, p = .026). We checked these results against R: rma(measure = "PLO", xi, ni) and predict(res, transf = transf.ilogit) in metafor 5.2, and the Freeman–Tukey version with transf.ipft.hm. The pooled estimates, CIs, τ², I² and Q agree to 6 or more significant digits.
Frequently asked questions
Which transformation should I use for a prevalence meta-analysis?
The logit transformation is a safe default and is recommended by recent methods papers. Freeman–Tukey is still common, but its back-transformation can mislead when study sizes differ a lot, so if you use it, check it against the logit result.
Can I include studies with 0% or 100% prevalence?
Yes. The logit and raw scales add 0.5 to the cases and non-cases of those studies (the metafor default), and the Freeman–Tukey and arcsine transformations work without a correction.
What does the prediction interval mean?
It is the range in which the true prevalence of a new, similar study is expected to fall, taking between-study heterogeneity into account. In prevalence reviews it is usually much wider than the confidence interval and is worth reporting.
Why is I² so high in my prevalence review?
Prevalence studies are often large and very precise, so even small real differences between populations give a high I². Look at τ², the prediction interval and the forest plot, and explore subgroups (region, diagnostic method, year).
Does this match R metafor or meta::metaprop?
It matches metafor’s rma() with the PLO, PFT, PAS and PR measures and inverse-variance pooling. meta::metaprop uses a GLMM by default; set method = “Inverse” there to get the same results.
Is my data uploaded?
No. All calculations run in your browser.
Do I need to log in or create an account to use the proportion meta-analysis calculator?
No. There is no login, no sign-up and no email required. Open the page and start using it straight away, for free, on any device.
How do I cite this tool in my thesis or paper?
Use the citation box on this page. In APA 7th edition: Tufail, M. A. (2026). Proportion meta-analysis calculator: pooled prevalence with forest plot [Online tool]. Codanics. Retrieved [date you used it], from https://codanics.com/tools/proportion-meta-analysis-calculator/. BibTeX, RIS (EndNote, Mendeley, Zotero), MLA, Harvard, Chicago and Vancouver formats are also available.
Free guides and courses from Codanics
More free tools
See all free Codanics tools → · Codanics on YouTube · WhatsApp channel

