I am fitting a valued Exponential Random Grpah Model (ERGM) of a bipartite network. I'm using R with packages ergm (from statnet), sand and network. I am using 'sum' as the valued term, the weights of the bipartite matrix as the response variable and some node attributes as covariates. I've fitted the model and obtained the estimates, but I want to know if there's a way to obtain the explained variance of each node attribute in the model.