Meta-analysis

Meta-analysis statistically combines effect sizes from multiple independent studies addressing the same question, producing a pooled estimate more precise than any single study could offer. It sits on top of a systematic review: the review finds and appraises the studies, the meta-analysis synthesises their numbers.

Each study's effect is weighted, usually by the inverse of its variance, so larger and more precise studies carry more influence. A fixed-effect model assumes one true underlying effect; a random-effects model allows the true effect to vary across studies and is the safer default in social-science research, where populations and designs differ substantially.

Heterogeneity is quantified with Cochran's Q and the I-squared statistic, then explored through subgroup analysis or meta-regression. Publication bias is examined with funnel plots and tests such as Egger's regression. A complete write-up reports the pooled effect, its confidence interval, heterogeneity and a sensitivity analysis.