What Is a Meta-Analysis? A Beginner's Guide

A meta-analysis is a statistical technique that combines results from multiple studies addressing the same question into a single, more precise summary estimate. Instead of reading ten trials and forming an impression, a meta-analysis mathematically pools their effect sizes — weighted by precision — into one number with a confidence interval.

Why pool studies at all?

Individual studies are often too small to detect a real effect reliably, or they show inconsistent results by chance. Pooling increases statistical power, narrows confidence intervals, and can reveal a genuine effect that no single study was large enough to confirm on its own.

What you need before you can run one

• A systematic review that has already identified and appraised the relevant studies.
• Comparable outcome measures across those studies (or a way to convert between them).
• Extracted effect sizes and variance/sample size data from each study.

You can't meta-analyze what you haven't systematically identified first — see our guide to systematic reviews for that earlier step.

The basic output

Results are typically displayed as a forest plot, with a pooled estimate (the diamond) summarizing the overall effect, alongside a heterogeneity statistic (I²) indicating how consistent the individual studies were.

When it's not appropriate

If included studies are too clinically or methodologically different (different populations, interventions, or outcome definitions), pooling them into one number can be misleading rather than informative. In that case, a narrative synthesis is the more honest choice.