Funnel Plot Maker for Meta-Analysis

You need a Study column, an Effect column, and one precision input: SE, Variance, or both Lower CI and Upper CI. Select the matching effect measure so OR, RR, and HR data are analyzed on the logarithmic scale.

Copy your study table from Excel and paste it into the generator, or upload a CSV or TSV file. The chart updates in your browser, so no R code or package installation is required before exporting SVG or PNG.

Treat asymmetry as evidence that effect estimates may vary with study precision, not as a diagnosis. Check heterogeneity, study methods, outcome selection, reporting practices, and chance before deciding which explanation is plausible.

No. Publication bias is one possible cause, but true effect differences, weaker methods in small studies, selective outcomes, and random variation can produce the same pattern. A contour-enhanced funnel plot adds context but still does not prove that studies are missing.

You can draw a funnel plot with fewer studies, but sparse patterns are difficult to interpret. Formal asymmetry tests usually have low power below about 10 studies. The generator therefore displays a caution when fewer than 10 valid studies are plotted.

Use standard error on the vertical axis for this meta-analysis funnel plot. The scale is reversed, so more precise studies appear near the top. Sample size can be used in other funnel-plot designs, but it is not interchangeable with standard error.

A contour-enhanced funnel plot shades two-sided significance regions around the no-effect value. The regions help you see whether gaps fall mainly in nonsignificant areas or elsewhere, but the visual pattern remains exploratory and must be interpreted with the study context.

A forest plot shows each study estimate and confidence interval, often with a pooled result. A funnel plot places estimates against precision to inspect small-study patterns. Use both: the forest plot for effect synthesis and the funnel plot for exploratory asymmetry assessment.