Hi all, I am using LGMs to assess if growth across 4 measurements can be predicted by an intervention (3 groups, modeled using dummies). I have 19 schools in my sample and I want to account for non-independence of the data.
I saw in a previous thread that it is recommended to use dummies when there are less than 20 clusters. I followed this advice and included dummies in my model, however this gives me the following error: WARNING: THE SAMPLE COVARIANCE OF THE INDEPENDENT VARIABLES IS SINGULAR. PROBLEM INVOLVING VARIABLE SCH7. NO CONVERGENCE. NUMBER OF ITERATIONS EXCEEDED.
Do you have any advice on how to solve the error? Thank you in advance!