Sara Vargas posted on Monday, October 25, 2010 - 6:30 pm
I am doing LGMs for numerous outcomes measured at 3 time points (baseline, 6 months, and 12 months). I am looking at differences in intercept and slope between the intervention and control groups. For many of the models, I allowed the 3rd time point to be estimated in order for the model to fit the data (e.g., not be set at 12 months). Now, I have been trying to re-center the intercepts at 6 and 12 months respectively in order to determine if there are signficant group differences at each time point. This gets messy with estimated time points. Is there another way to get this information? Thank you!
You can use Model parameter labels and introduce New parameters in Model constraints, where the New parameters are the means at the different time points for the two groups. You express these means in line with the formulas that we have in our Mplus Topic 3 short course handout.