I've been struggling with the selection of analytic approach to a longitudinal study (two time points) where we are interested in examining the development of a symptom in a clinical sample, and the contribution of various phenomena to this development, and interrelationships between several symptoms. The study design is shaped in such a way all the concepts are measured at both time points in the same sample, with a year in between. I understand that growth curve analysis usually requires three or more data collection waves, and that the two time points impose some restrictions on what analyses can be performed. From some intense googling and and reading of relevant literature, I have come across Latent Change Score modelling, which in theory seems to be a valid to examine our hypotheses. I can't seem to grasp just what separates this approach from latent growth curve analysis, and whether these approaches are in any case apt for our aims. If anyone here can provide some clarity or guidance, this very junior researcher would be very pleased!