Independent subgroup contrast in meta-analysis
= Independent subgroup contrast in meta-analysis
{title2=$Z=(\widehat\mu_1-\widehat\mu_2)/\sqrt{v_1+v_2}$}
For <independent> subgroup estimates $\widehat\mu_1,\widehat\mu_2$ with <variances> $v_1,v_2$, the null contrast has <Wald statistic> $Z=(\widehat\mu_1-\widehat\mu_2)/\sqrt{v_1+v_2}$. If evidence overlaps, include minus twice the <covariance> in the contrast variance. A nonsignificant comparison is not proof of equivalence.