Asparouhov, T., & Muthén, B. (2003). Full-information maximum-likelihood estimation of general two-level latent variable models with missing data.
(2) I saw Dr. Bengt Muthén give a presentation at APS in May of 2005. In that presentation, he talked about a longitudinal latent variable model where the indicators were changing over time -- either people aged out of an old measure or a measure was updated. There was a period of time where both measures were given, and an attempt at linkage was made. I haven't successfully been able to find a reference to this procedure. Is it possible to obtain a reference or a copy of that training?
(2) I talked about that in our Johns Hopkins training which was recorded and is on our website together with handouts. See handout for Topic 4, slides 98-99. I have tried to inspire people to investigate this further and write about it, but it hasn't been done yet as far as I know. An application of this thinking is in
Pettit, G.S., Keiley, M.K., Laird, R.D., Bates, J.E. & Dodge, K.A. (2007). Predicting the developmental course of mother-reported monitoring across childhood and adolescence from early proactive parenting, child temperament, and parents’ worries. Journal of Family Psychology, 21, 206-217.