\usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \sqrt {.4} $$\end{document} + .08 * M)2. If the correlation between T and M ran via A, however, then the genetic variance of T was calculated as .15 + (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \sqrt {.25} $$\end{document} + .08 * M)2, where .15 is associated with the cross-path relating T and M.Table 5Results Simulation study 4: false negative rates under the extended univariate moderation model and the full bivariate moderation model when the covariance between T and M is not moderatedDrop β a Drop β c Drop β e % hitsnsim% hitsnsim% hitsnsimT and M correlated via A Ext univariate.201999.311998.482000 Full bivariate.071998.212000.502000 p < .001 p < .001 p = .47T and M correlated via C Ext univariate.271999.161999.472000 Full bivariate.252000.031998.482000 p = .18 p < .001 p = .95T and M correlated via E Ext univariate.481999.471999.911997 Full bivariate.251998.281989.912000 p < .001 p < .001 p = .83 Note: β a, β c and β e denote the moderation parameters on the variance components unique to T (see Fig. 1). nsim denotes the number of data sets (out of 2000) for which the extended univariate moderation model and the
