Hello. I'm running a generalized linear model with binomial error structure (looking at survival data). I'm interested in looking at the relationship between the predicted response variable and one of the covariates; however, to do this, do I need to run the model without the covariate in which I am interested (as I'll be graphing the response against the raw predictor variable)? It was just my understanding that its inappropriate to graph the predicted values for the full model against one of the raw predictor variables, but I just wanted to make sure. Any help you can provide would be much appreciated. Thanks!