I'm running a LGMM looking at distress levels over the four years of college with distress measured twice a year, once for each semester. There appears to be seasonal effects where people report being less distressed in the spring then in the fall. While this isn't surprising, I'm not sure how to measure this effect in a LGMM framework. Do you have any suggestions?
This can be handled in a couple of ways. Piecewise growth modeling can be used with time broken up into seasons. If there aren't enough (3-4) time points per piece, an extra growth factor can be introduced for even just one time point. And mixtures can be introduced. For the use of the last 2 ideas see the paper on the Mplus web site (under Papers, GMM):
Greenbaum, P.E., Del Boca, F.K., Darkes, J., Wang, C. & Goldman, M.S. (2005). Variation in the drinking trajectories of freshman college students. Journal of Consulting and Clinical Psychology, 73, 229-238
For a related paper, see also Beets et al. (2009) in Drug and Alcohol Dependence.