Meta-analysisSystematic review
A study of studies: pooling results statistically to get a more precise answer than any single trial gives.
A systematic review searches for every study that addresses a question using pre-declared criteria; a meta-analysis then combines their results statistically into a single pooled estimate. Together they are the standard way to summarise a field rather than cherry-picking from it.
It matters because individual studies disagree, and a review that declares its search and inclusion rules in advance cannot quietly select the flattering ones. It also reveals how much the studies disagree — heterogeneity — which is often more informative than the pooled number.
In practice, a meta-analysis is only as good as its inputs: pooling ten small, poorly-controlled trials produces a precise-looking estimate of a badly-measured effect. Read what was included, not just the headline. Publication bias — unflattering results never being published — pushes pooled estimates in a predictable direction.
Sources
Related terms
Written and reviewed by Mathew Beale, MSc Biotechnology, University of Reading.
Last reviewed: