I apologise if this is a really stupid question, but I am finding it really difficult to work out which type of analysis to use. I have two large datasets from 1995 and 2015 that are cross-sectional (from the same area, identical sampling methods, but not the same participants). I want to look at the change in categorical data across the two time periods - specifically the change in support network types controlling for age, marital status, education, length of residence etc. I have run a multinomial logistic regression using 'year of study' as an IV, but I don't think that this is correct. Is there a model I can use in Mplus that is more appropriate?
I assume that "support network types" is a nominal DV so that multinomial logistic regression makes sense. Then a dummy covariate captures the 1995 vs 2015 year difference. The slope for that dummy - which varies across the nominal categories is what orchestrates the change in category probabilities, so I think you are on the right track.