Hi,
I want to conduct a meta-SEM to test the predictors and outcomes of a variable.
My question is: I don’t have the correlations between each independent variable (IV) and other IVs, nor between each dependent variable (DV) and other DVs, since I extracted them from different studies. Given this, how can I construct a correlation matrix? Can I set the correlations between IVs or DVs to zero?