Evidence-appraisal glossary
Tau-squared
Tau-squared is the estimated variance of the true effects across studies in a random-effects meta-analysis, measured on the same scale as the effect. Its square root, tau, is the estimated standard deviation of those true effects.
Also called: between-study variance, tau2.
It is the between-study variance that widens the confidence and prediction intervals in a random-effects model, and unlike I-squared it is an absolute measure of how much the true effects spread. The caveat is that it is estimated imprecisely when only a handful of studies are pooled, and different estimation methods can return noticeably different values from the same data.
This is a plain-language methodology definition for reading research. It is general education, not medical advice.