hello there!

I'm running a mixed effects analysis. I have 2 continuous covariates, 1 categorical factor (with 4 levels) and all possible 2 and 3-way interaction terms in my model. I plotted my categorical factor in seperate lines, one of my continuous covariates on the x-axis and my other continuous covariate in seperate plots (as mean-1 SD, mean and mean+1 SD, i.e., 3 different plots). I used the default covariates labeling (which is "Labels"). But then I noticed that using other covariates labeling options change the intercepts of my 3 plots.

How come different covariates labeling options change the plots (especially intercepts) without making any changes on values in the estimates table?

I've read the following descriptions of the 3 types of labeling options, but could not really get the difference among them, and how they could change my plots:

"Covariates labeling decides which label should be associated with the estimates and plots of simple effects as follows:

Labels produces strings of the form Mean±SD

Values uses the actual values of the variables

Labels+Values produces labels of the form Mean±SD=XXXX, where XXXX is the actual value."

thanks in advance!

tutku